Nominating the President : Evolution and Revolution in 2008 and Beyond 9780742566392, 9780742566378

Citation preview

NOMINATING THE PRESIDENT

NOMINATING THE PRESIDENT Evolution and Revolution in 2008 and Beyond

Edited by JACK CITRIN AND DAVID KAROL

ROW M A N & L I T T L E F I E L D P U B L I S H E R S, I N C. Lanham • Boulder • New York • Toronto • Plymouth, UK

ROWMAN & LITTLEFIELD PUBLISHERS, INC. Published in the United States of America by Rowman & Littlefield Publishers, Inc. A wholly owned subsidiary of The Rowman & Littlefield Publishing Group, Inc. 4501 Forbes Boulevard, Suite 200, Lanham, Maryland 20706 www.rowmanlittlefield.com Estover Road Plymouth PL6 7PY United Kingdom Copyright © 2009 Rowman & Littlefield Publishers, Inc. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior permission of the publisher. British Library Cataloguing in Publication Information Available Library of Congress Cataloging-in-Publication Data

Nominating the President : evolution and revolution in 2008 and beyond / edited by Jack Citrin and David Karol. p. cm. Includes bibliographical references and index. ISBN 978-0-7425-6637-8 (cloth : alk. paper) — ISBN 978-0-7425-6639-2 (electronic) 1. Presidents—United States—Nomination 2. Presidents—United States— Election—2008. 3. United States—Politics and government—21st century. I. Citrin, Jack. II. Karol, David, 1969– JK521.N665 2009 324.273'15—dc22 2009018161 Printed in the United States of America

⬁ ™ The paper used in this publication meets the minimum requirements of American National Standard for Information Sciences—Permanence of Paper for Printed Library Materials, ANSI/NISO Z39.48-1992.

CONTENTS

Acknowledgments

vii

Introduction Jack Citrin and David Karol 1

2

1

The Changing Nature of Campaign Financing for Presidential Primary Candidates Richard L. Hasen

27

Money in 2008: A Collapse of the Campaign Finance Regime? Thomas E. Mann

47

3

Known Unknowns in Campaign Finance Allison R. Hayward

4

Assumptions and Realities of Presidential Primary Front-Loading Andrew E. Busch

77

What Political Scientists May (or May Not) Know about Presidential Nominations Barbara Norrander

95

5

6

The Rules and the Role of Race and Gender in 2008 Jennifer A. Steen

v

57

109

vi

7

Contents

The Way We Choose Presidential Nominees: Problems and Prospects David E. Price

141

References

149

Index

159

About the Editors and Contributors

167

ACKNOWLEDGMENTS

his volume assesses the state of the presidential nominations process in the United States. The chapters by leading scholars of party politics consider how developments in campaign finance, the campaign calendar, and the strategies of candidates shaped the remarkable 2008 contests in both major parties. The authors look to the future by asking whether 2008 represents a revolution in the process or just an evolutionary change in a fundamentally stable system. This book is the outcome of a conference titled “Nominating the President: The Evolving Process and Its Effects” sponsored by the Institute of Governmental Studies and the School of Law at the University of California, Berkeley, on April 11–12, 2008. We are deeply grateful to Dean Christopher Edley of the Berkeley School of Law and to the Harris Fund of the Institute of Governmental Studies for financial support. The revised versions of the conference papers published here have benefited greatly from the presentations and comments of the distinguished faculty, former public officials, journalists, and political consultants who participated in the conference panels. We gratefully acknowledge the contributions of Terri Bimes, William Brandt, Christopher Edley, Gina Glantz, Peter Hart, Ken Khachigian, Goodwin Liu, Susan Rasky, Eric Schickler, Lynn Sweet, Robert Wickers, John Yoo, Byron York and John Zaller. We also thank Raymond Wolfinger for his assistance in making the conference a success and we recognize the special effort of U.S. Representative David E.

T

vii

viii

Acknowledgments

Price to attend the event as the keynote speaker. Thanks also to John Hanley and Andrew Kelly for valuable research assistance and to Ethan Rarick and Maria Wolf of the Institute of Governmental Studies for their work in organizing the conference and in preparing the manuscript.

INTRODUCTION Ja c k C i t r i n a n d D a v i d K a r o l

he American system for nominating presidential candidates is unique in its openness and complexity. Over two centuries the process has evolved, in an episodic series of punctuated equilibria, toward greater democratization. The main rules of the modern game, the McGovern-Fraser reforms, were established in 1972 in the aftermath of a bitter internecine conflict among the Democrats and were designed to increase the power of rank and file voters in the selection of the presidential nominees.1 The remarkable 2008 campaign is the catalyst for this inquiry about continuity and possible new changes in the nomination process. Developments in campaign finance, the scheduling of primaries and caucuses, the role of superdelegates, and the ability of candidates to mobilize support among distinct social groups influenced the outcomes of the process and may shape future elections.

T

CHANGE AND CONTINUITY 2008 was a year of unexpected and sometimes unprecedented developments. The early verdict on the impact of the post–1968 reforms was that they would usher in an era where more ideologically extreme or “outsider” candidates would prevail over the choice of the party establishment and where the media rather than party leaders would be “kingmakers” (Polsby 1983). The vagaries of “momentum” generated by media coverage of candidates’ fortunes in a handful of early states could determine outcomes (Bartels 1988). 1

2

Introduction

However, more recent scholarship suggests that by 1980, the party elites had regained control of the steering wheel and were guiding the process toward the selection of a candidate satisfactory to the main constituents of their electoral coalition. Between 1980 and 2004, in an “invisible primary” party elites usually converged on the eventual winner before the first vote was cast in a caucus or primary election (Cohen, Karol, Noel, and Zaller 2008). Not in 2008. For the first time since 1972, a front-runner was upset as Barack Obama defeated Hillary Clinton for the Democratic presidential nomination. In the Republican Party, for the first time since 1964 there was no clear front-runner as the season of primaries and caucuses began. In the end, John McCain, an erstwhile insurgent whose candidacy was declared moribund by many observers late in 2007, won the nomination. In both parties, voters witnessed the most competitive and prolonged process in a generation. Even after the Iowa caucus and New Hampshire primary, both nomination contests were wide open. Finally, history was made, as the two leading candidates in the Democratic Party were an African American and a woman. On the surface at least, 2008 seemed to presage less centralized direction of the selection process and an expansion and loosening of party coalitions as wealthy activists and a plethora of bloggers weighed in. There were other dramatic developments in 2008. Voter participation was higher than ever, at least on the Democratic side. Internet fundraising, already growing in past years, exploded and gave small donors a greater role in the process than they have had since records have been kept. At the same time, the voluntary system of public financing created in the early 1970s appeared to have finally collapsed, with all the leading candidates in both parties opting out of it and the eventual president doing so even in the fall campaign. The Internet emerged as a powerful force in raising money and communicating about the campaign. Finally, the effort of many states to move up in the primary calendar led to protracted conflict over the seating of the delegations of two large states, Florida and Michigan, a battle that arguably damaged the chances of the presumptive Democratic frontrunner, Hillary Clinton. This was the first major battle over seating delegates since the 1972 Democratic convention in which McGovern forces barred Mayor Da-

Introduction

3

ley’s Illinois delegation and fought off challenges to the legitimacy of the California delegation. Yet if there were many dramatic departures from the norm in 2008, there was also visible continuity. The clearest case was the stability in the laws and rules determining the structure of the competition. The 2008 calendar varied little in key respects from prior years as both national committees granted state parties and governments considerable discretion in scheduling their primaries and caucuses. Once again, instead of the national primaries that are increasingly employed abroad and which American survey respondents have long favored, state parties chose delegates on different dates in a process extending over five months. Enormous attention was once again focused on Iowa and New Hampshire and, to a lesser extent, South Carolina. The constraints of the calendar were similar to the effects evident in previous election years. Serious candidates were forced to organize well before the beginning of the voting in Iowa on January 3, 2008. Without an incumbent president or vice president running, the field of candidates in both parties was large. But as in the past, some candidates did not survive the “invisible primary” phase and dropped out before the first vote was cast in the Iowa caucuses (see table 0.1).2 As Barbara Norrander shows in chapter 5, contests in the early states once again mattered a great deal. Candidates who skipped these early contests or fared poorly in them were swiftly eliminated, leaving a handful of contenders to compete in the remaining states. Although the start of the nominations marathon began earlier than ever and more delegates were allocated in the first stages of the race than ever before, these changes merely continued the “front-loading” trend that began in the 1980s and has been manifest in each succeeding cycle (Mayer and Busch 2004.) State parties persisted in allocating delegates through a mix of primaries, caucuses, and even old-fashioned state conventions. While Republicans distributed delegates through procedures ranging from winnertake-all at the statewide level to proportionality, Democrats remained bound by the post–1968 rules and only awarded delegates to candidates in proportion to their share of the popular vote. We suggest later in this chapter that the rules for allocating delegates were a contributing factor in determining the outcomes in both parties.

4

Introduction

Table 0.1.

Casualties during the Invisible Primary: The Day a Candidacy Died They Tested the Waters but Didn’t Go Ina

October 12, 2006 Mark Warner (D-Va.), former governor November 12, 2006 Russ Feingold (D-Wis.), Wisconsin senator November 29, 2006 Bill Frist (R-Tenn.), U.S. Senate majority leader December 2, 2006 Tom Daschle, former U.S. Senate majority leader December 10, 2006 George Allen (R-Va.), U.S. senator December 15, 2006 Evan Bayh (D-Ind.), U.S. senator January 24, 2007 John Kerry (D-Mass.), U.S. senator September 8, 2007 Chuck Hagel (R-Neb.), U.S. senator September 29, 2007 Newt Gingrich (R-Ga.), former speaker of U.S. House of Representatives Pre-Iowa Dropouts February 23, 2007 Tom Vilsack (D-Ia.), former governor June 14, 2007 Jim Gilmore (R-Va.), former governor and RNC chair August 12, 2007 Tommy Thompson (R-Wis.), former governor and HHS secretary October 18, 2007 Sam Brownback (R-Kan.), U.S. senator December 20, 2007 Tom Tancredo (R-Colo.), U.S. representative a

Two widely discussed potential candidates were Gen. Wesley Clark and former Vice President Al Gore. Both avoided making absolute rejections of candidacy, but neither ran. Of these two, Clark seemed to be testing the waters more seriously.

The rules governing campaign finance were also largely unchanged from 2004. The last major change in campaign finance regulation, the Bipartisan Campaign Reform Act (or McCain-Feingold), dates from 2002 and did not have significant implications for presidential nomination campaigns. As in prior years, a voluntary system of partial public financing was available to candidates who qualified for it and were willing to accept the required limitations on their expenditures. What differed in 2008 was the innovative use of the Internet by the Obama campaign to raise an unprecedented level of funding. This fundraising effort is likely to leave an enduring imprint on the choices of strong candidates in future elections and was itself discussed at length by the media as evidence of the strength of Obama’s campaign. Finally, observers have persistently argued that opening up the nominations process to rank and file voters would divide the parties. Scholars reasoned that faction-based candidates would engage in a lengthy campaign of mutual recriminations. As a consequence, intraparty competition would weaken the ability of the parties to unite for

Introduction

5

the general election campaign (Polsby 1983). This fear of internal conflict echoed loudly among Democrats as the competition between Obama and Clinton continued until June, with a crescendo of calls on the senator from New York to concede defeat beginning long before the final primaries in early June. Yet in the current period of partisan polarization and more ideologically cohesive parties, both Democrats and Republicans were able to coalesce around their respective nominees in the fall. The Democrats set aside a long and bitter primary campaign that primed race and gender cleavages, while John McCain overcame the long-standing mistrust of many conservative Republicans toward him, in part by choosing Sarah Palin as his running mate.

POLITICAL PARTIES IN THE NOMINATIONS PROCESS The first two presidential elections were nonpartisan in character; George Washington faced no real opposition. Subsequent contests, with the sole exception of the 1820 election,3 have seen real competition, almost always dominated by the nominees of two major parties. Once partisan conflict became institutionalized, the nominations process evolved in the direction of increasing popular participation. Originally, caucuses of each party’s members of Congress chose their presidential and vicepresidential candidates. After Andrew Jackson—seemingly the people’s choice—was denied office in 1824, this system, derided by critics as “King Caucus,” lost legitimacy and gave way to the national conventions as the arena where nominations were made. The move to conventions was democratizing, as control shifted from several dozen elected officials to several hundred and later to thousands of state and local party leaders and activists. The most common system for selecting delegates to the national conventions resembled today’s caucuses. Party loyalists would meet in localities to elect delegates to a county meeting; these delegates might then attend a district convention that elected delegates to a state convention; the state convention would then select the delegation to the national convention that actually chose the party’s nominee. However, by contemporary standards, participation in this process was limited, and local party leaders could often dominate the delegate selection process.

6

Introduction

In the early 20th century, primaries were adopted by many state legislatures, giving voters a greater say in the selection of delegates. However, most delegates were still not chosen in primaries. Not all primaries offered voters real choices either. Candidates were selective in entering them, often deferring to “favorite sons,” typically a state’s senator or governor who sought to bring a delegation nominally pledged to themselves to the convention as bargaining chips. Until the post–1968 reforms, the most important role of primaries was to gauge a candidate’s appeal. When internationalist Republican Wendell Willkie entered the Wisconsin primary in 1944 to show that he could win votes in an isolationist state and lost badly, his campaign was over. By contrast, when Catholic Senator John F. Kennedy entered the West Virginia primary in 1960 hoping to show strength in a rural, Protestant state, his triumph paved the way to his party’s nomination. It can be argued that in 2008 Obama’s strong showing in Iowa and New Hampshire sent a strong signal that he could win in states with few black voters. In the past, however, the important recipients of the primary voters’ message were party leaders unsure of whom to support. Today what matters as well, if not more, is how ordinary voters, guided in part by the media, use the outcomes of early primaries and caucuses to assess the viability and relative strength of the contenders. Before 1972, it was possible for candidates with strong support from party leaders to be nominated without entering or winning primaries, as the cases of Illinois Governor Adlai Stevenson in 1952 and Vice President Hubert Humphrey in 1968 illustrate.4 By 2008, the idea that party leaders could set aside the apparent verdict of the primary campaign bordered on sacrilege. Fighting to frame themselves as the candidate of “the majority,” Clinton and Obama quarreled about how to interpret the number of contests each had won and the meaning of their respective vote totals. The legitimacy of having the superdelegates—party leaders and elected officials—cast the deciding votes was hotly debated. The highly competitive nature of the contests in 2008, especially on the Democratic side, meant that familiar aspects of the system loomed larger because it seemed possible that they would affect the outcome, something that was not true of contests in either party for a generation. For this reason, there was much discussion of the fairness of the

Introduction

7

nominating rules over the course of the long campaign. This was not because the rules determining the allocation of delegates among candidates were new. To the contrary, all the important provisions had been in place for several cycles at least. But in 2008 the contests in both parties were competitive enough for the rules relating to the delegate selection process (primaries vs. caucuses), the delegate allocation rule (proportionality vs. winner-take-all), and the role played by unpledged “superdelegates” in the Democratic Party to matter.

THE LESSONS OF 2008 Front-Runners Can Lose The first few nomination contests after the McGovern-Fraser reports yielded surprising results by conventional standards. In 1972 George McGovern, who understood the strategic imperatives of the rules of the new system he had helped design, upset the early favorite, Senator Ed Muskie.5 McGovern appealed to anti–Vietnam War liberals but had difficult relations with many other traditional Democratic constituencies, including southerners, northern machines, organized labor, and Jews. McGovern’s nomination split his party and resulted in spectacular defeat. Four years later Democrats nominated a more moderate candidate, Jimmy Carter. Like McGovern, Carter was not a favorite of the party establishment. Yet by dint of his personal campaigning in Iowa he won a surprise victory and parlayed that into media coverage, which generated momentum that led to his success in later contests. This was a remarkable victory. McGovern was at least a modestly familiar figure to national observers, whereas Carter had been almost totally unknown. The Democrats’ McGovern-Fraser reforms also affected Republicans, partly because several state legislatures were controlled by Democrats and created primaries for both parties. As a result, delegates elected in primaries and open caucuses predominated in GOP as well as Democratic conventions from 1972 onward. This shift had no immediate effect in 1972 when President Nixon was easily renominated. Yet the fruits of reform were visible in the campaign for the Republican presidential nomination in 1976 when President Gerald Ford barely survived

8

Introduction

a challenge from former California Governor Ronald Reagan. This contest went all the way to the convention. It was the last time a convention opened with any real uncertainty as to the outcome.6 While Reagan’s campaigning ability and appeal to conservatives far outshone that of the unelected president, the overwhelming support Ford received from party and elected officials suggests that he would have had far less trouble securing renomination under the pre-reform system. As noted above, initial response to the consequences of the postreform system by scholars and pundits was sharply negative. The most prominent academic critic, Nelson W. Polsby (1983), contended that the system undermined the party elites who had traditionally provided “peer review” of presidential aspirants. Polsby held that in place of the traditional tendency of conventions to pick experienced, “safe” candidates who had had broad appeal throughout their party, the new process favored “factional” candidates like McGovern and Reagan who enjoyed intense support from only one wing of their party. These candidates would be less likely to win and less likely to govern effectively. Updating Polsby’s analysis, however, Cohen and colleagues (2008) conclude that even in the postreform system “the party decides.” Candidates have learned that they need to enter the fray early and contest every primary and caucus. This increases the incentive to win support in the form of endorsements and campaign contributions, which in turn buttresses the power of elected officials and interest group leaders. Party leaders monitor each other’s moves and assess the ideological reliability and electoral viability of contending candidates. From 1980 on, both parties ultimately chose the candidates most palatable to the mainstream. By the standard criteria, Hillary Clinton won the Democrats’ invisible primary in 2008. Although Obama was hardly an unknown, having been anointed as a “comer” through his selection as the keynote speaker at the 2004 national convention, Clinton was more well-known than Obama, she had collected more endorsements and money, and she was deemed the likely Democratic nominee in the public opinion polls. Her ultimate defeat indicates both that party elites do not always converge on a winning candidate and that new developments in political mobilization have made the boundaries of the political party, if it is defined as a coalition of officials and interest groups with intense policy concerns, more permeable. In a system where rank and file voters have influence,

Introduction

9

those who inform their preferences—bloggers, talk show hosts, and the like as well as party dignitaries—can be important. And this poses risks for front-runners. The self-proclaimed maverick John McCain also seemed an unlikely choice as the consensual pick of the Republican coalition. True, McCain had made strenuous efforts to gain the acceptance of social conservatives (Cohen, Karol, Noel, and Zaller 2008), but when the voting began in Iowa he lagged behind several others in the fragmented Republican field in both money and standing in the polls.

Caucuses Matter Caucuses are an older system than primaries. In recent decades it is chiefly the Iowa caucuses that have attracted attention because of their status as the first contest in the nomination process. Sometimes the intenwse coverage of this event gives the impression that the caucus is some quaint local ritual unique to Iowa. In fact, many other states use caucuses. In 2008, for instance, in addition to Iowa, 11 states as well Guam and American Samoa selected delegates to the Democratic National Convention via caucuses. On the Republican side, 12 state parties in addition to Iowa’s used caucuses or similar processes, as did the territorial parties of American Samoa, Guam, the Northern Marianas, and Puerto Rico. The importance of the success of Obama in caucus states and the Clinton campaign’s disastrous neglect of these contests has been widely discussed, with the obvious implication that future candidates will not repeat Clinton’s error. The Republican race in 2008 also showed that whether a state party uses a primary or caucus can matter greatly for how well a candidate fares. But in this case, the favorite of caucus voters did not prevail. While he was still running, Mitt Romney won every caucus except the one he really needed to: the one in Iowa. Mike Huckabee also did relatively well in caucuses, most famously in Iowa. The former Arkansas governor also prevailed in the Kansas caucuses and virtually tied McCain in Washington state. Ron Paul, who had intense appeal to a loyal, if limited, following, also did notably better in caucuses—at least relatively speaking—than in primaries. Most of Paul’s double-digit showings were in caucus states. By contrast, the eventual GOP nominee, John McCain, struggled in caucuses.

10

Introduction

Caucuses invariably draw far fewer voters than primaries. This is true even of the Iowa caucuses, by far the best attended of all. The caucus system advantages candidates who have strong local organizations or passionate supporters. There is also some reason to believe that better educated voters are more likely to be overrepresented in caucuses. Since passion is seldom associated with moderation, there is reason to believe that caucus attendees are more extreme in their views than primary voters and this can hurt the more centrist, front-running candidates that the invisible primary (and actual primaries) tend to favor.7

Winner-Take-All Can Help (or Hurt), and So Can Proportionality While in most states both parties held either a primary or a caucus, the methods of delegate allocation that Democrats and Republicans used within the same states differed notably in ways that favored some candidates and harmed others. The Democrats are rigid. Regardless of whether delegates are chosen by a primary or a caucus, they are allocated to candidates on a proportional basis closely reflecting their support in the state. By contrast, Republicans allow state parties great leeway. While Republican parties in several states use the winner-take-all system for allocating delegates, this is just one option and it is not exclusively chosen. There are also state Republican parties that embrace compromise plans. For example, California Republicans traditionally used a statewide winner-take-all system that awarded the candidate who won the most votes in the GOP primary (even if he won far less than a majority) all of the state’s delegates. In 2008, however, California Republicans adopted a winner-takeall system at the congressional district level.8 Other state Republican parties, including those in New York and New Jersey, moved in another direction and switched to the statewide winner-take-all plan. The delegate allocation system had important consequences. Table 0.2, which reports the allocation of delegates in Missouri, illustrates this fact. Remarkably, Missouri’s voters divided their support almost equally among both remaining Democrats and the three major Republican candidates still running on Super Tuesday. Yet the consequences of the different rules the parties employ were starkly visible in the delegate totals candidates received from the Show Me State. Obama’s lead over Clinton was microscopic, and the two Democrats split the delegation evenly.

Introduction

11

Table 0.2. Different Rules, Different Results: The Case of the Missouri Presidential Primaries

Democrats Obama Clinton Edwards Uncommitted

Votes

Delegates

Republicans

Votes

405,284 (49%) 395,287 (48%) 16,747 (2%) 3,135 (1%)

36

McCain

36

Huckabee

194,304 (33%) 185,627 (32%) 172,564 (29%)

0

Romney

0

Paul

Delegates 58 0 0 0

By contrast, McCain’s very narrow lead over Romney and Huckabee yielded him the entire Missouri delegation. The case of Missouri offers perhaps the most dramatic demonstration of the consequences of the partisan divergence in delegate allocation rules, but similar results were evident in many other states. In general, the proportional system favored Obama because he won large majorities in small states (frequently via caucuses) and large minorities in populous ones. Clinton won most of the big states, as her campaign accurately trumpeted, but these victories did not give her a huge payoff in terms of delegates because of the Democrats’ employment of proportionality. In the Republican race, however, McCain, unlike Clinton, was handsomely rewarded for his first-place finish in large states. One way to clarify the effect of the delegate allocation rules is to conduct a “what if ” test and calculate the number of pledged delegates the candidates would have won under alternate arrangements. It might be argued that this counterfactual analysis is a problematic exercise, because under different rules candidates would have employed different campaign strategies. This is doubtless true, but as Jennifer Steen reveals in chapter 6, candidates did consistently appeal to different demographics. Under any electoral system it seems likely that Clinton would have had more appeal than Obama to the Latinos of Texas and the white working class of West Virginia, while Obama would have outdrawn her among African Americans in Mississippi and white liberals in Vermont. Similarly, Mike Huckabee’s appeal to socially conservative voters was a fundamental feature of the Republican race unlikely to be affected by the choice of winner take all or proportionality. .

12

Introduction

Figures 0.1 and 0.2 report the pledged delegate totals won by the candidates and those under an alternative set of rules. Figure 0.1 compares the actual pledged delegate totals for the Democrats (with Florida and Michigan excluded) and the simulated totals on the assumption that the candidates’ performance in the states were unchanged but that delegates were allocated on a statewide, winner-take-all basis. This comparison shows that the proportionality rule worked strongly to Obama’s advantage. Had states allocated delegates on a winner-take-all basis Clinton would have had a lead in pledged delegates larger than the one Obama had under the current system. The handful of delegates John Edwards won before dropping out—delegates who might have been pivotal had the race been closer still—would not have existed under the winner-take-all system. For the Republicans, too, the rules affected the outcome. The comparison presented in figure 0.2 is slightly different from that in fig-

Figure 0.1. Pledged Delegate Allocation to Democratic Presidential Candidates under Current Rules and under the Winner-Take-All Rules Republicans Allow.

Introduction

13

Figure 0.2. Pledged Delegate Allocation to Republican Presidential Candidates through Super Tuesday 2008 under Current Rules and under Democratic Proportional Rules.

ure 0.1. Unlike the case for the Democrats, the Republican race did not really continue through all of the states. After Super Tuesday, John McCain was widely seen as the inevitable nominee. Although Mike Huckabee remained in the race for a month, Mitt Romney withdrew. So what is compared is the actual distribution of delegates after the Super Tuesday results and the hypothetical distribution had Republicans always used the proportionality rules of the Democrats. Under the proportionality scenario, John McCain would have been in a far weaker position and no candidate would be close to a majority. Indeed, under the proportionality system, it seems likely that Romney would have stayed in the race after February 5 and that no candidate would have secured a majority of delegates before the Republican convention.

Superdelegates Can Count but Don’t The role of superdelegates and the legitimacy of their voting power was a novel cause of controversy among Democrats in 2008, again because the Democratic race was so competitive. The gap between Clinton

14

Introduction

and Obama among pledged delegates was sufficiently narrow that the superdelegates were pivotal. In a system dedicated to the effective participation of rank and file party members, a determinative role for elected politicians and often obscure party officials can be problematic. For this reason the Democrats have pledged a “significant reduction” in the number of superdelegates in future contests.9 Nevertheless, the superdelegates seem likely to survive in some form beyond 2008. Politicians tend to prefer the institutional arrangements under which they have been elected. President Obama’s path to the White House was not seriously impeded by the superdelegates. As the underdog, Obama initially received fewer endorsements from superdelegates than his rival Hillary Clinton, but he fared much better with the late-deciding delegates as the race developed and he took the lead among pledged delegates. The prospect of a primary challenger to Obama in 2012 seems remote, but as the incumbent he would expect to benefit from superdelegate support if a challenger emerged. Since presidents traditionally dominate their parties’ national committees, any change in the process would probably require his approval. Moreover, being a superdelegate is one of the few perks of membership in the Democratic National Committee. So while some DNC members reportedly did not relish the pressures they came under during the 2008 campaign, it is not clear that most would be eager to give up this privilege even if Obama left the decision to them.10 Yet while the institution of superdelegates seems likely to remain, whether they can play a meaningful role in the nominating process is unclear. One widely debated question is what the creators of the superdelegates intended. In chapter 7, U.S. Representative David Price, who served as staff director of the Democratic National Committee panel that established the superdelegates, offers a minimalist interpretation of their intended role. Price rejects claims that the superdelegates were created to be “kingmakers.” In his view, the idea was that the superdelegates were to function as a kind of insurance, available to provide leadership to conventions in case a prospective nominee was eliminated by death, illness, or scandal. Price also contends that it was seen as desirable for the party to have members of Congress present at the national convention, as had been routine before the McGovern-Fraser reforms, and that the superdelegate system would assure this.

Introduction

15

Yet many other observers have had a more expansive understanding of the superdelegates’ role.11 In this alternative to Price’s perspective, the superdelegates were supposed to take an active role in the process. Free from the passions of caucus-goers who do not reflect the rank and file and better informed than primary voters, the party leaders, whose status was validated by their having won elected office, would be able to guide the process toward the selection of a more pragmatic, electable choice, much like the bosses who gathered in the fabled smoke-filled rooms of conventions past. The McGovern-Fraser reforms were instituted in part because the system in which party leaders—not voters—chose conflicted with widespread mistrust of authority and calls for participatory democracy. The immediate catalyst, of course, was the nomination of Hubert Humphrey in 1968 even though he did not compete in a single state’s primary. In the context of the intense disagreement then raging within the Democratic Party over the Vietnam War, the view that the old system had lost legitimacy was widespread. As Price (1984: 148) notes, “The tumultuous Chicago convention of 1968 confirmed the insurgents’ sense of the illegitimacy of party operations and also left many regulars convinced of the need for fundamental change.” When the superdelegates were created in 1982, the older “mixed system” was still remembered by many party leaders and activists and, in the aftermath of the McGovern and Carter nominations, looked better to many in retrospect. It is true that in the first contest after their creation the superdelegates played a meaningful role without too much controversy. In 1984, superdelegates heavily favored the front-runner, former Vice President Walter Mondale, over his leading opponents, Senator Gary Hart and Rev. Jesse Jackson. Superdelegate backing allowed Mondale to effectively capture the nomination after the end of the primary season in June and avoid a messy period of haggling before or during the convention. Yet Mondale had won more voter support than either of his leading rivals, albeit less than their combined total, and because of this his opponents could not rouse much indignation about the superdelegates’ role.12 If such a multicandidate scenario were to recur, perhaps the superdelegates could once again bring closure to the process by favoring the leader without facing great legitimacy concerns. Yet this is a very limited discretionary role for the superdelegates and one unlikely to recur with regularity.

16

Introduction

There is little reason to believe that the legitimacy problems of leadership control have diminished in the ensuing 40 years. It seems unrealistic to think that superdelegates could really pull the trigger and block the leader among pledged delegates without generating a backlash that made their action counterproductive. A contest in which the superdelegates become relevant would be one in which at least two candidates retained strong support throughout the primary season. In 2008, the nightmare scenario for many Democrats was that a charismatic African-American candidate would be barred from the nomination by party officials despite winning more support from the voters than any other candidate. Many feared that in such a case party divisions would linger, hurting Democrats’ chances in the fall and beyond. Hypothetically, the party could have been able to coalesce around a ClintonObama ticket brokered by superdelegates. In general, just as in 2008, uncertainty about the potential fallout of intervention is likely to deter superdelegates from trying to redirect the process even when this is technically within their power. There are other constraints on the likelihood that superdelegates will choose to play a decisive role in the nominations process. One obstacle is the collective action problem familiar to social scientists. While far more influential than an ordinary voter or caucus participant, each superdelegate still has only a minute impact on the overall outcome, possessing only one vote out of 4,418 in 2008. The superdelegates realize that their individual actions are unlikely to affect the party’s choice. But they also know that making public their preference for a candidate before the race is settled can greatly affect their relationships with the eventual nominee and potential president. Accordingly, superdelegates, like other politicians, have a strong incentive to “bandwagon” or side with the candidate they think is likely to be nominated, even if he or she is suboptimal from an electoral or governing standpoint. The superdelegates’ dilemma is especially acute if a candidate they fear would hurt the party’s chances in November faces more than one opponent. In such a case, superdelegates and other party leaders face a coordination problem as well; it may be unclear who should be the focal point to rally behind in order to prevent nomination of the unwanted candidate. This problem was evident in 2004 prior to the Iowa caucuses. There was much concern among leading Democrats about the

Introduction

17

allegedly unelectable frontrunner for the nomination, Howard Dean, but no clear alternative among his rivals had yet emerged. As it happened, the voters solved this problem for Democratic elites by voting for John Kerry, the more mainstream choice. Yet it is easy to imagine a messier scenario in which Dean fared better initially while one of his rivals did well in Iowa and another emerged in New Hampshire. Collective action and coordination problems have always plagued parties during the nominations process. As Cohen and colleagues (2008: 94) note, a bandwagoning dynamic was evident in pre-reform contests as well. In the pre-reform era, however, these problems were less acute. There were party leaders who controlled large blocs of votes; as late as the 1960s entire state delegations were often under the sway of a governor, senator, or other party boss who could act as a kind of super superdelegate and because there were fewer actors each with greater influence than any single superdelegate today, it was easier for them to converge.

Campaign Finance Evolves The two signal developments in campaign finance in 2008 were the great increase in small donations, largely via the Internet, and the flight from the federal system of voluntary public financing. The growth in small contributions has been widely seen as a positive development. As Richard Hasen, Thomas Mann, and Allison Hayward show in this volume, the rise of small donors has the potential to narrow the gaps in thinking between advocates of greater campaign finance regulation and skeptics. Yet while striking, both of these major developments in campaign finance in 2008 are best understood as the culminations of trends rather than as sudden and unexpected changes. The role of online donations first drew comment in 2000 when John McCain and Bill Bradley gained attention for raising funds in this manner.13 The sums these candidates raised pale in comparison to those collected by candidates in later years, however. In 2004 Howard Dean raised far more on the Web than either McCain or Bradley had. Dean’s Internet operation helped to boost him to the position of fundraising leader among the Democratic contenders, a remarkable achievement for an obscure former governor of a tiny state.

18

Introduction

In 2008, three candidates raised substantial sums in small donations— Barack Obama, Hillary Clinton, and Ron Paul. Similarly, the abandonment of the public financing system was not a sudden or unexpected development, but rather the logical conclusion of an ongoing trend. In earlier years candidates with great fundraising ability or massive personal wealth had sometimes opted out of the public financing system with its limits on state-level spending. Republicans John Connally in 1980 and Malcolm “Steve” Forbes Jr. in 1996 are examples. Yet no candidate had rejected public financing and won the party’s nomination until 2000 when George W. Bush, a champion fundraiser, opted out and succeeded anyway. On the Democratic side, no candidate even attempted to opt out until Howard Dean did so in 2004. Democratic hopefuls may have been deterred by fears that liberals and the media would react negatively to a Democratic candidate who abandoned a symbol of limiting the power of the wealthy. Yet given his enormous resources due to successful fundraising, Dean decided to take the plunge, despite the fact that his core constituencies, highly educated white liberals, were presumably the strongest supporters of campaign finance reform. The absence of a backlash to Dean’s move convinced John Kerry that he could follow suit and rely on his personal funds. With this recent example, Obama and Clinton were free to escape the limits of public financing, with Obama’s campaign arguing that his reliance on so many relatively small donors was a form of democratic participation. For all the talk of developments in fundraising and regulation, the 2008 campaign also reminds us that in politics money is not everything. It is a necessary component for success, but far from a sufficient one. John McCain won his party’s nomination despite being outspent by both Mitt Romney and Rudy Giuliani. If we put aside the large sums Romney contributed to his own campaign, Giuliani was actually the champion fundraiser on the GOP side, yet he wound up with only one pledged delegate. By contrast, Mike Huckabee had little success raising money, yet won the Iowa caucus and captured first place in seven other states. Similarly on the Democratic side, Hillary Clinton prevailed in the late primaries in Pennsylvania, Ohio, and Texas despite being substantially outspent by Barack Obama in these contests.

Introduction

19

WHITHER PARTY CONTROL? Early critics of the McGovern-Fraser reforms decried the weakening of party elites and warned about the power of media and momentum (Polsby 1983; Bartels 1988). But after 1980, the pendulum appeared to swing back toward party control. No longer were parties nominating outsiders or extremists. Instead, after a period of learning about the properties of the new system party elites had adapted and were steering the process again (Cohen et al. 2008). The failure of elites in either party to both unite behind a candidate before Iowa and see that candidate nominated in 2008 calls the party revival thesis to the bar of new evidence. This is especially true when one also recalls the disarray of Democrats in 2004. While John Kerry was the early front-runner going into that race, he stumbled in 2003 as Howard Dean drew the spotlight. Kerry only emerged as Dean’s chief challenger because of his win in Iowa. Moreover, absent the funds he was able to draw upon from his personal resources, the Massachusetts senator’s campaign might not have survived until Iowa. While Kerry, an experienced war hero without a strong ideological edge, was exactly the kind of candidate who would have risen to the top in the pre-reform era, the way he actually got the nomination was messier than proponents of the party revival thesis would have predicted. There were events in the 2008 cycle that suggested the continued vitality of party organizations as constraints that shape the incentives of candidates, even if not as cohesive establishments that select the nominee. As in previous campaigns, prominent candidates adjusted their positions on key issues to appeal to important constituencies in their parties (Karol 2009). For all the symbolic freight of his candidacy, the policy differences between Obama and his chief rival, Hillary Clinton, were quite limited. Both candidates were broadly acceptable to all key interest groups aligned with the Democratic Party. Moreover, while Obama was in some ways an insurgent candidate, he had already forged connections to leading Democrats during his brief time in Washington and was not seen as threatening core party constituencies. This was in sharp contrast to the troubles Gary Hart had with organized labor or McCain had with the religious right, the NRA, antitax activists, and other groups in 2000.

20

Introduction

Still, there is no gainsaying that the nominations of both Obama and McCain are an empirical challenge to the thesis of party revival and control. Whether or not 2008 portends a major change in the constraints of the nominations process, though, is uncertain. For one thing, elite control of the nominations game after 1980 was never absolutely certain or complete. Some of the “establishment” candidates like Walter Mondale and Robert Dole who did prevail did so only with difficulty and luck. Conversely, we may observe that Obama’s victory over the establishment’s early favorite was a very narrow one. As Andrew Busch points out in chapter 4 on front-loading, inferences about the enduring properties of the system we study are inevitably based on a small number of cases and as a result must always remain tentative. Why, then, did Democratic elites fail to both identify a favored candidate and lead that candidate to victory in 2004 and 2008? Two other explanations are plausible. First, rather than the consequence of a transformed system, the 2004 and 2008 contests could be described as “context-specific”: the intrusion of a disruptive issue, in this case the war in Iraq, may have changed the nature of the process in these two cycles. When there is no salient issue that divides a party and galvanizes an intense minority faction, it is easier for the organizational advantages and signaling behavior provided by party leaders to direct outcomes. This seems to be the story of Al Gore’s defeat of Bill Bradley in 2000 and the elder President Bush’s triumph over Bob Dole in 1988. Yet when an issue such as Iraq divides a party it is much harder for elites to coalesce and carry the day. In three of the four messy Democratic nominations in the postreform era—1972, 2004, and 2008—the party was deeply divided over responses to a war. On the other hand, one way in which the properties of the system may be changing in an enduring way is the impact of new communications technology on campaigning. The ability to raise large sums and coordinate volunteers via the Internet was crucial for the insurgent candidacy of Howard Dean and still more dramatically for Barack Obama’s challenge to the presumptive nominee. Some candidates have always been able to raise large sums in small donations and build campaigns that did not depend on the usual big donors. Previous fundraising methods such as direct mail and later 1-800 numbers were far more costly for campaigns and generated less revenue

Introduction

21

than working through the Web. The Web may have destabilized parties, at least temporarily weakening the hold that party elites had reestablished in the postreform era. Historically, candidates relying on small donations from the grass roots tended to be polarizing outsiders (e.g., George Wallace, Pat Robertson, Jerry Brown) with no real chance of winning the nomination. But Web-based funding may permit such candidates to more easily qualify for public funds and stay in the game longer. Clinton and Obama were unusual in being candidates with mainstream appeal who could raise substantial sums in small amounts. Whether this is a result of their being trailblazers or a function of the potential of the Web is unclear at this point. Future candidates, including front-runners, will certainly try hard to maximize their Web-based fundraising and mobilizing. However, it is worth noting that the only Republican to raise substantial sums online in 2008 was Ron Paul—a figure much more in the mode of traditional recipients of small donations. Therefore, generalizing from the cases of Clinton and Obama may be hazardous. And neither division over the war in Iraq or the autonomy afforded by Web-based fundraising can explain the failure of Republican elites to coordinate in 2008. The party was united on the war and its nominee did not rely on Internet fundraising.

CONCLUSION Undoubtedly the most dramatic aspect of the 2008 campaign was the prominence of the first African-American and first female candidates who were seen as viable presidential contenders. As Jennifer Steen points out in chapter 6, there was no change in the rules that made the emergence of a credible female or African-American candidate more likely in 2008 than in previous contests. Undoubtedly, over the past 40 years, public attitudes have become far more conducive to the emergence of such candidates, as evidenced by the steady increase in the number of survey respondents telling pollsters that they would be prepared to vote for otherwise qualified candidates who happened to be African American or female. General Colin Powell also enjoyed enormous popularity in the 1990s and would have been a strong candidate for either party’s nomination.

22

Introduction

When Hillary Rodham moved to Arkansas with Bill Clinton in 1973, there were no women governors, senators, cabinet secretaries, or Supreme Court justices. Since then things have changed dramatically. By 2007, 22 states had elected at least one woman as governor. The share of congressional seats occupied by women has increased in every Congress since 1980. In short, a female or African-American presidential nominee, while virtually unimaginable in 1960, might well have emerged several years ago. The critical element in 2008 was the openness of the Democratic race and the availability of two very strong candidates in a favorable electoral context. Looking toward 2012, while some tinkering with the rules seems inevitable, it is unlikely that there will be fundamental changes in the nature of the nominations process. The atypically prolonged and competitive fight for the Democratic nomination in 2008 did highlight several aspects of the rules that had long escaped scrutiny. As a result, the Democratic National Convention voted to create a committee, the “Democratic Change Commission.” The group was appointed in March 2009 and instructed to report by 2010. Its charge was to propose changes in the rules governing the superdelegates, the scheduling of contests, and the conduct of caucuses. In each area, the directions to the Change Commission indicated a desire for solidifying the democratization of the nomination process. The number of superdelegates was to be significantly reduced. The schedule was to be reviewed with the goals of holding the line on the beginning of the contest and avoiding a recurrence of conflicts over seating delegations, as occurred in the cases of Florida and Michigan in 2008. The caucuses were to be made more accessible by better notification about their meeting sites and providing for absentee voter participation. While Democrats typically have squabbled more than Republicans about the nominations process, at present there is more potential than usual for change on the GOP side. This is the case because, as Andrew Busch notes in chapter 4, in 2008 Republicans empowered the Republican National Committee and not only the national convention (as had been the case previously) to modify the rules. Moreover, in contrast to the Democratic side, there is reason to expect a competitive race for the Republican nomination. With no obvious choice, the rules of the game could matter more. In the absence of a sitting president with effective control of the

Introduction

23

party organization, the current chair of the Republican National Committee had greater discretion in appointing the Temporary Delegate Selection Committee, which is also scheduled to report in 2010. That body’s proposals will require approval of two-thirds of the Republican National Committee to be adopted. While President Obama will have the last word on the Democratic process, greater controversy is likely among Republicans. Advocates for various 2012 hopefuls doubtless will try to turn the calendar and other rules to their advantage, making consensus on major change difficult to achieve. Thus the basic outlines of the process—a staggered series of events with great diversity across state and local parties in terms of procedure— seem certain to remain intact. There is no reason to believe that either party is about to make a radical change like banning caucuses. There is also no serious discussion of the national primary that voters have long favored. Even regional primaries seem like a remote prospect. Historically, the main stimulus for reform has been the belief that the existing rules hurt the party’s chances in the general election. And there is no widespread sentiment among Republican elites that the current process produced a poor nominee; most think McCain was the Republicans’ best hope in a very uphill campaign. Moreover, the massive Republican defeat in 2008 should be a spur for building rather than weakening the party’s cohesiveness and organizational strength. The Democrats have traditionally been more inclined to fine-tune their rules, especially after electoral defeat. But 2008 was a year of great triumph for Democrats, thereby reducing the pressure for reform. So even if the number of superdelegates is reduced, it is highly unlikely that they will be abolished, and their impact is quite limited in any case. Public financing in presidential nomination campaigns does seem doomed without some major reform that increases candidates’ incentives to participate in the system. Relaxing state spending limits and making the matching funds more generous might fit the bill, but it is not clear that such a reform is in the offing. The fact that Richard Hasen and Thomas Mann, two longtime advocates of public financing, are, in chapters 2 and 3, quite sanguine about recent developments in the process and the rise of small donors in particular suggests that the impetus for more reform in the campaign finance system has greatly diminished.

24

Introduction

CODA Even if the party and governmental rules and regulations surrounding presidential nominations evolve only slightly, that does not mean that adaptation and strategic innovations on the part of candidates will not occur. One lesson of recent elections is that institutional changes are not necessarily the most profound factors in shaping outcomes. Generalization about presidential nominations is not easy. The process is complicated and the number of cases scholars have to observe is small. Aggregation of cases is achieved only over time, and from cycle to cycle the process is evolving. Certain events are sufficiently numerous and certain rules sufficiently predictable in their effects—the impact of winnertake-all is one example. Yet the exciting if unexpected developments of 2008 remind us that the process is fundamentally a stochastic one. There is no single indicator that is an infallible indicator of victory. The historical record shows that leading in early polls, raising the most money, bagging the most endorsements, winning Iowa and New Hampshire are all correlates of success, but no candidate in either party accomplished all of these feats in 2008. Prediction remains a difficult business, especially about the future.

NOTES 1. These rules changes are often referred to as the McGovern-Fraser reforms because U.S. Senator George McGovern and U.S. Representative Donald Fraser cochaired the commission set up by the Democratic Party to review the rules in the wake of the disorder at the 1968 Democratic National Convention in Chicago. 2. The first group listed in the table did not formally announce their candidacies (although Senator Bayh set up an exploratory committee with the FEC). 3. In this election, which occurred during the Era of Good Feelings, during which one of the two parties had collapsed, President James Monroe faced no opponent. 4. This was no longer the norm, however. Reiter (1985) shows that in the last decades of the system before the McGovern-Fraser reforms, nominations seldom were multiballot affairs. Favorite sons became less numerous, and winning candidates usually competed in at least some primaries. Reiter attributes

Introduction

25

these trends to social change and the decline of traditional party organizations at the local level. See also Carleton (1957). 5. One of the secrets of McGovern’s success was the strong organization in caucus states that his opponents lacked, the same technique that paid off for the Obama campaign in 2008. The Clinton campaign had far less excuse than McGovern’s rivals for failing to contest these states. 6. Some might cite Senator Ted Kennedy’s last-ditch attempt to repeal the “robot rule” at the 1980 Democratic convention freeing pledged Carter delegates to turn to him. Yet while it provided some drama and heightened tensions, this maneuver was seen as a long shot even at the time. 7. Unfortunately, there are very few polls of caucus participants and most of these focus on the comparatively well-attended and atypical Iowa caucuses, so political scientists do not know as much about who attends caucuses as we would like. 8. Since the distribution of Republican candidates’ support differed relatively little across regions within states, in practice winner take all at the district level yielded results very similar to those a statewide winner-take-all plan would have produced: McCain won the lion’s share of the delegates. 9. “Parties Altering 2012 Calendars,” The Hill.com (accessed March 25, 2009). 10. Members of the Republican National Committee are also unpledged automatic delegates. Since they make up a much smaller share of the total convention vote than their Democratic counterparts and since no GOP contest has been as competitive as the Democratic races in 2008, they have never mattered and their existence has sparked almost no discussion. 11. Price acknowledges that at least some commission members did see the delegates as a way to increase the role of traditional party insiders. 12. Ironically, Mondale, the choice of the superdelegates, was wiped out in the fall, carrying only his home state of Minnesota. 13. “Fundraising Champion Bush Lags Online,” AP State and Local Wire, February 4, 2000; “Low-Cost Fund-Raising on Web Aids Candidates,” Boston Globe, February 7, 2000.

1 THE CHANGING NATURE OF CAMPAIGN FINANCING FOR PRESIDENTIAL PRIMARY CANDIDATES R i c h a rd L . H a s e n inancing a presidential nomination campaign has been a complicated affair for some time: candidates face contribution limits, source limitations, extensive reporting requirements, and, for those candidates opting into the voluntary public financing program, a strict spending limit. Moreover, the campaign financing system is not only complex but dynamic. The system recently has undergone two fundamental changes: (1) the voluntary public financing system, with its accompanying strict spending limit, is no longer an option for serious presidential candidates, who would be grossly outspent by nonparticipating rivals; and (2) the campaigns have come to rely ever more upon campaign finance bundlers, who collect maximum contributions (currently set at $2,300) from friends and associates, and upon “microdonations,” donations of under $200. These changes arose because of the changing legal rules governing money in presidential elections and a dramatic lowering of the costs of campaigning facilitated by the Internet. They may alter both the nature of presidential campaigns as well as the types of candidates who are viable to become general election candidates. This chapter has three main parts. First, it describes the two recently emerging fundamental changes in the campaign financing system for presidential primary candidates for the major parties.1 Second, it explains why these changes have emerged now. Finally, it evaluates whether these changes are likely to be positive or negative from the standpoint of political equality and in terms of the quality and type of presidential candidates who win their parties’ nominations.

F

27

28

Chapter 1

CAMPAIGN FINANCE RULES AND DYNAMIC CHANGE Under rules created by the Federal Election Campaign Act (FECA) in 1974, individuals could donate up to $1,000 during the primary season and another $1,000 for the general election to presidential candidates. The Bipartisan Campaign Reform Act of 2002 (BCRA, more commonly known as McCain-Feingold) raised the FECA limits to $2,000 for each election, and the figure was $2,300 per election for 2008, thanks to an inflation adjustment mechanism built into BCRA. Federal law also imposes source limitations on contributions: corporations, labor unions, most foreign citizens, and some other entities are barred from contributing to federal candidates. Candidates must report information about contributions and spending to the Federal Election Commission (FEC) in periodic reports.2 Presidential candidates also may decide to opt into a public financing system created by FECA. Candidates during the nomination season who opt in become eligible for public financing by raising $5,000 or more in contributions of $250 or more in at least 20 states. Participating candidates then have the first $250 of additional contributions matched by the government. In exchange, participating candidates agree to cap their overall total spending in the primary election season.3 The amount of the cap is set by a formula in FECA that Congress did not touch in BCRA. In 2004, the cap was $50.4 million; in 2008 the cap was about $52 million4 (Green 2006: 95; Federal Election Commission 2008). Major party nominees may choose whether or not to participate in a separate public financing system for the general election, a choice not dependent upon whether they participated as publicly financed candidates during the nomination process.5 Except for a handful of self-financed candidates, all serious major party candidates accepted public financing during the primary season until the 2000 election.6 In that election, George W. Bush, the eventual Republican nominee, opted out of the system, raising $95.6 million in privately funded donations. Al Gore, the eventual Democratic nominee in that election, accepted matching funds and raised contributions up to the public financing spending cap of $49.4 million (which included $15.5 million in public financing) (Malbin 2006: table 11.1). In 2004, both of the parties’ eventual nominees opted out of public financing

The Changing Nature of Campaign Financing for Primaries

29

during the primary season. Each raised contributions many times the $50.4 million cap that would have applied to publicly financed candidates. Kerry raised $234.6 million; Bush raised $269.6 million (Malbin 2006: table 11.1). By 2008, it was clear that the public financing system, with its relatively paltry spending limits, was a luxury no serious candidate could afford, at least in the primary season. Indeed, in 2008, the only candidates opting for public financing were the weak ones who could not raise funds well on their own. John Edwards agreed to take public financing as his personal fundraising collapsed (Kirkpatrick 2007), and John McCain agreed to take it when he too was written off as having no chance at securing the Republican nomination. Once McCain became the front-runner, he backed out of his commitment to take public financing in the primary election season (Mosk 2008). During the primary season (up to each party’s national convention), eventual Democratic Party nominee Barack Obama raised $414.2 million through private contributions,7 and eventual Republican Party nominee John McCain raised $216.3 million, without taking any public financing (Campaign Finance Institute 2008a: table 2.) In one month alone, July 2008, Obama raised $50 million and spent $55 million, spending more in one month than he would have been able to spend for the entire primary campaign season had he opted for public financing and its accompanying $52 million spending cap (Kuhnhenn 2008; Daily Kos 2008). Figure 1.1 shows the campaign contributions (and in Gore’s case, a share of public financing) raised by the eventual nominees from the major political parties in the 2000, 2004, and 2008 primary seasons. It is not only the eventual winners who have been raising a great deal of money. The total money raised by all candidates running in the major party nominating contests rose from $332.7 million in 2000 to $673.8 million in 2004, to nearly $1.2 billion in 2008 (Malbin 2006: table 11.1; Campaign Finance Institute 2008a: table 28). Adjusted for inflation, the 2000 figure would be about $425 million and the 2004 figure about $785 million. See figure 1.2. Besides a general increase in the amount of money raised by candidates during the primary season over the last three elections, two key fundraising trends have emerged in recent elections: bundling and microdonations.

30

Chapter 1

Figure 1.1. Total Contributions Raised by Eventual Major Party Nominee During Primary Season.

Bundling. Beginning with George W. Bush’s 2000 election, major party candidates began asking some of their contributors to solicit their friends, acquaintances, and business associates to donate to the candidate. These volunteers, who usually ask for the maximum legal individual donation, then “bundle” checks from their circle. Those bundlers reaching a certain fundraising threshold receive access to the candidate and other perquisites from the campaign. In 2000, 226 Bush “Pioneers” each raised at least $100,000 for the campaign (in $1,000 maximum individual contributions), giving Bush over $22 million in bundled funds9 (Green and Bigelow 2006: 58). In 2004, when the individual contribu-

Figure 1.2. Total (Inflation-Adjusted) Dollars Raised by Candidates in Major Party Primary Contests, 2000–2008.

The Changing Nature of Campaign Financing for Primaries

31

tion limits doubled from $1,000 to $2,000, Bush raised at least $77 million through bundling, about 30 percent of his total donations (Green 2006: 104). Kerry, the Democratic nominee, raised at least $41.5 million through bundling, about 18 percent of his total donations (Green 2006: 105). In the 2008 campaign, as of August 18, 2008, 534 bundlers raised at least $75.75 million for McCain, and 509 bundlers raised at least $63.3 million for Obama (Center for Responsive Politics 2008). Figure 1.3 shows the sum of the minimum amounts of bundling activity during the primary season by the major party candidates who eventually secured the nominations of their political parties. Microdonations and Microdonors. The other notable campaign finance trend in the campaign finance process is a dramatic increase in both the total amount of donations given to campaigns in amounts under $200 (microdonations) and the number of donors giving $200 or less to a presidential primary campaign in the aggregate (microdonors). It is important to disentangle two threads here because they have different normative implications. Let’s begin with microdonations, remembering that under these definitions, a person who donates $150 twice to a candidate during an election has given two microdonations but is not a microdonor. Here is what we know about microdonations. Such donations have made up an ever larger share of presidential nomination process fundraising in the last three elections. Microdonations accounted for 25 percent of the total contributions to major party candidates seeking

Figure 1.3. Summed Minimum Amounts of Bundled Contributions Raised During Nomination Season by Eventual Major Party Nominees.

32

Chapter 1

Figure 1.4. Percentage of Total Contributions Given in Amounts under $2000 (Microdonations) to Major Party Presidential Primary Campaigns, 2000_2008.

their parties’ nomination in the 2000 election season. The figure rose to 34 percent in 2004 and 38.8 percent in 2008 (Malbin 2006: table 11.1; Campaign Finance Institute 2008a: table 210). See figure 1.4. It is unclear if there is an emerging party divide on leading candidates’ reliance on microdonations.11 In 2000, Al Gore raised 20 percent of his total primary funds and George W. Bush raised 16 percent of his total primary funds from microdonations. In 2004, John Kerry raised 37 percent of his total primary funds from microdonations.12 George W. Bush raised 31 percent of his total primary funds in this manner (Malbin 2006: table 11.1). Thus, before 2008, there appeared to be some parity in the reliance of the major party candidates on microdonations. In 2008, however, Barack Obama raised more than half (53 percent13) of his primary funds from microdonations.14 In contrast, John McCain raised 31 percent of his funds (over $62 million) from microdonations. (Campaign Finance Institute 2008a: table 2). Figure 1.5 shows the total amounts of money raised by eventual party nominees during the primary season from microdonations since 2000. Though Obama raised much more in microdonations in 2008 than Bush did in 2004, there was much more parity when it came to the number of microdonors, that is, people who gave no more than $200 in the aggregate to the primary campaign. The range of reliance on microdonors among the top major party candidates in the last two presidential primary election seasons was 20–26 percent, with Obama coming in at 26 percent, Bush (2004) at 25 percent, McCain at 21 percent,

The Changing Nature of Campaign Financing for Primaries

33

Figure 1.5. Total Dollars Raised During During Primary Season in Amounts Under $200 (Microdonations) by Eventual Party Nominees.

and Kerry at 20 percent (Campaign Finance Institute 2008c; see also Mosk and Cohen 2008; S. Cohen 2008.) See figure 1.6. Yet because Obama’s overall fundraising was so much greater than other candidates, he raised a much larger total amount of money in the primary period from microdonors than Kerry (2004), Bush (2004), or McCain (2008). See figure 1.7 (relying on data from Campaign Finance Institute 2008c). Obama raised $117.7 million from microdonors in the primary, Bush $64 million, Kerry $43.1 million, and McCain $42.9 million. With these statistics in hand, we turn to two sets of questions. Why have these changes occurred? And what do these changes mean

Figure 1.6. Percentage of Total Dollars Raised From Microdonors (Giving Under $200 in the Aggregate) by Eventual Party Nominees.

34

Chapter 1

Figure 1.7. Total Millions of Dollars Raised from Microdonors (Giving Under $200 in the Aggregate) by Eventual Party Nominees.

for our democracy about the nature of presidential campaigns and candidates?

EXPLAINING THE CHANGES This section considers why the patterns of campaign financing in presidential nominating processes have changed. We begin with consideration of the demise of the public financing system, then turn to the candidates’ increasing reliance on bundlers and microdonations.

The Demise of Public Financing The public financing system collapsed primarily because of a combination of legal and market forces. The public financing system enacted first in 1976 imposes a spending cap that has not kept up with the ability of candidates to raise money on the private market. Bush’s fundraising during the primary season in his first election, in 2000, showed that private fundraising efforts could almost double the amount available to candidates during the primary season. Bush raised $95.6 million privately, while Gore’s spending was capped at $49.6 million (Malbin 2006: table 11.1). (Note that the “primary season” runs until the summer nominating convention, so that once a party candidate becomes the presumptive nominee, he or she may continue to raise funds and then spend them attacking the other party’s candidate during that period.)

The Changing Nature of Campaign Financing for Primaries

35

Bush’s innovation increased candidates’ demand for political money during the primary season. Knowing that the sky was now the limit, candidates had an ever increasing need for funds, if only not to be outspent by an opponent. But the supply of money has expanded as well. In 2002, Congress (through BCRA) doubled the amount that an individual could give to federal candidates. In 2004, one-third of all contributions to presidential primary candidates came from donors giving at least $2,000. Bush, the eventual Republican nominee, received 44 percent of his contributions from this group; Kerry, the eventual Democratic nominee, received 24 percent from this group (Malbin 2006: table 11.1). In the 2008 primary season, 25.6 percent of contributions came from contributors giving at least $2,300 (the new maximum, thanks to BCRA’s inflation adjustment). McCain, the eventual Republican nominee, received 33 percent of his contributions from this group. Obama, the eventual Democratic nominee, received 15 percent from this group during the primary (Campaign Finance Institute 2008a: table 215). Once Bush in 2000 showed how much more money he could raise beyond the low spending caps, the arms race began, and it received extra impetus from the increased supply of larger money that wealthier donors could now give thanks to BCRA. Though Democrats had something of an ideological commitment to the public financing system, in 2004 both the eventual Democratic nominee Kerry and early front-runner Howard Dean opted out. In 2008, both the eventual Democratic nominee Obama and early front-runner Hillary Clinton opted out. On the Republican side, McCain appears to have shifted his decision on opting in and opting out with his political fortunes; only when it looked like his campaign was failing did he express an interest in opting in, a position he reversed when he recovered politically. At this point, the public financing system is dead for presidential primaries unless it is reformulated by Congress.

The Rise of Bundling, Microdonations, and Microdonors Candidates seeking a major party presidential nomination have always pursued multiple fundraising routes, with one of the most lucrative being the $1,000 per plate (or now $2,300 per plate) dinner with

36

Chapter 1

the candidate. With 500 attendees, such a dinner can clear $1 million. But as presidential candidates need to raise ever more money and to travel across the country in search of primary votes and not just money, other means of raising money have become attractive as well. In this environment, the use of campaign bundlers has become central. Bundling shifts some of the onus of fundraising from the candidate to the candidate’s supporters. Campaign bundlers likely are motivated by a number of reasons, including the desire to help elect a candidate of choice. But some bundlers likely are looking to curry influence with the candidate, securing perquisites such as access to the candidate at special events. Bundling became more attractive to individuals looking for access after the 2002 passage of BCRA, which outlawed the giving of six- and seven-figure “soft money” donations to the political parties. Soft money used to be the entry price for access to presidential candidates; bundling to some degree has replaced that activity. Not everyone is capable of being a bundler. Only those people with a circle of friends or associates who can cough up maximum individual contributions (currently $2,300) can raise funds in this way. Bundling also creates a sense of competition; the publication of bundling activities likely spurs some competitiveness among bundlers for the same candidate, with different levels of participation for those who reach higher bundling levels. Microdonation targeting arose out of a combination of candidates’ increased demand for money and the reduction in fundraising costs thanks to the Internet and, in the 2008 election, cell-phone text messaging. Internet campaigning already was important in the 2004 election (Magleby and Patterson 2006: 169; Corrado 2006: 28–29). But its use appears to be accelerating rapidly. The rise of “cheap” political speech (Volokh 1991) has increased the ability of voters to exchange political messages via e-mail, instant messaging, amateur videos posted on YouTube and related sites, and more frequently, social networking sites such as MySpace and Facebook (Pew Research Center 2008). Text messaging is the next frontier. The Obama campaign sent 2.9 million text messages in one day to supporters who signed up to be “among the first” to hear of Obama’s vice presidential pick (Schatz 2008). The campaign then used the collected e-mail addresses and cell-phone numbers to follow up with potential contributors.

The Changing Nature of Campaign Financing for Primaries

37

Most importantly for our purposes, the cheap speech of the Internet allows the candidate to quickly and cheaply communicate with supporters, asking many people to give small amounts of money. Once the campaign invests in the fixed costs of Internet communication, the marginal costs of sending out numerous solicitation e-mails are small. And supporters can be hit repeatedly. One donor interviewed by the Los Angeles Times gave $700 to Obama over a period of months in $25 increments (Morain 2008). Repeaters were a critical part of the Obama fundraising strategy in 2008. The Campaign Finance Institute (2008c) reports that over 200,000 Obama donors started off giving contributions of $200 or less and then crossed the reporting threshold by giving more: “About 93,000 of these repeaters gave in cumulative amounts of no more than $400 for the full primary season. Another 106,000 repeaters ended up between $401 and $999.” Future studies should consider why these donors gave repeatedly. How much was due to enthusiasm for the candidate, how much to the long campaign, and how much to the lower transaction costs of repeat giving over the Internet with saved account information? The Obama story is useful in seeing how campaign fundraising techniques are additive rather than substitutionary. Obama did not choose to replace bundling with Internet-based small fundraising; he aggressively pursued both.

The Current Status of Presidential Primary Fundraising At this stage of the process, Republicans have relied more heavily on maxed-out donors and bundlers than Democrats have. Democrats have relied more heavily on microdonations. See figure 1.8, showing the disparity in the 2008 primary election season.16 But candidates from both parties have pursued both bundling and microdonor strategies, and most likely will continue to do so. With the supply of money increasing because of relaxed contribution limits and cheaper Internet-based communications, and the demand of candidates apparently insatiable due to the “arms race” of the nominating process, the amounts of money likely to be raised for nomination races should continue to rise. Before turning to a normative evaluation of this trend, it is worth noting one more potential effect of campaign finance law on future fundrais-

38

Chapter 1

Figure 1.8. Percentage of Money Raised in 2008 Presidential Primary Season by Eventual Major Party Nominees Through Bundling and Through Microdonations.

ing strategies. The Supreme Court under Chief Justice John Roberts has begun striking down a host of campaign finance regulations as potential violations of the First Amendment. It is conceivable that within one or two more presidential elections, the Court will strike down limits on contributions to political candidates (Hasen 2008a). If the Court took that step, one would expect to see candidates shifting at least some of their fundraising attention to “superdonors” giving six- and seven-figure donations. This strategy would not completely displace bundling and the targeting of microdonors, but it would play a central role in major campaigns.

A NORMATIVE EVALUATION OF CAMPAIGN FINANCE TRENDS The earlier sections of this chapter established that the campaign finance system for the presidential nomination process has undergone significant changes in recent elections and that both law and market forces have caused these changes. We turn now to the normative question whether these changes are positive or negative. In asking this question, we must first establish a normative baseline. For example, if one is concerned with the total amount of money spent on politics, then the trends emerging from the most recent presidential election nominating contests is especially bad: to-

The Changing Nature of Campaign Financing for Primaries

39

tal funds raised by eventual nominees during the primary election process appear to be rising, and overall spending by the candidates during the primary season certainly is rising significantly, from $332.7 million in the 2000 election to well over $1 billion in the 2008 election. However, total money spent on politics is not a good measure of the health of our campaign process. Instead, let us consider the normative question from the point of view of political equality and candidate attributes and quality.

Political Equality Is the current campaign finance regime in the primary season more or less egalitarian than in the past? The question is hard to answer because of crosscurrents: the demise of public financing has harmed the goal of political equality while the rise of microdonors has helped it. The public financing system, when it still worked, marginally promoted “barometer equality” (Hasen 2003: 111). Under this view, campaign spending should roughly mirror public support for the candidate’s political ideas. The presidential public financing matching fund program for primaries followed this model to some extent by matching the first $250 of each campaign contribution received by a candidate. It made small donations twice as attractive as they otherwise would be and in that way created an incentive for campaigns to go after donors contributing relatively small amounts of money. Of course, the system’s effects on political equality were limited: still only a small pool of people gave money to presidential candidates, and those donors tended to be wealthier than the general population. The legal and market forces described in the last section of this chapter caused the public financing system in the primaries to become obsolete. Moreover, the doubling of individual contribution limits in 2002 has since attracted campaigns to relatively wealthier contributors who could max out their contributions at $2,000 or more. Both of these forces have the tendency to make the system less egalitarian, making larger donations more important and smaller donations less important. However, these factors have been counteracted by the meteoric rise of microdonors. Microdonors promote barometer equality: the greater the number of microdonors supporting a candidate and the more intensely they do so, the greater the funding for the candidate to

40

Chapter 1

get out his or her message to the people. It is astounding that between the primary and general election, Obama was able to raise $150 million from over 1 million individuals giving no more than $200. Though Obama’s fundraising may be unique to the enthusiasm surrounding his candidacy, he is not the only presidential candidate who has had impressive microdonor fundraising. George W. Bush in 2004 raised $64 million from microdonors. Even McCain, whose fundraising generally lagged behind the other eventual nominees since 2000, raised nearly $42 million in 2008 from microdonors (Campaign Finance Institute 2008b). These trends are promising from the point of view of more Americans getting involved and helping to support their preferred candidates for president in a meaningful way. This is not to say that the current system is a fully egalitarian one. Even in 2008, about one-fourth of all money raised by presidential primary candidates came from $2,300 maxed-out donors. Moreover, most eligible Americans still have not donated to political campaigns, and many are too poor to give even modest amounts to campaigns. If the class of donors differs in important ways from nondonors, then the campaign finance system might not be promoting political equality but instead may be skewed in one direction or another. The concern that donors are unrepresentative of most Americans is one that merits further consideration. A study of donors in the 2004 presidential election found that donors who gave more than $200 were “more highly educated, older, and much wealthier than most Americans” (Institute for Politics, Democracy, and the Internet 2006: 15). However, “The small donors [i.e., microdonors, giving under $200] stand somewhere between the large donors and the general public in many respects. Small donors were neither as wealthy nor as highly educated as large donors, although they were more so than the general population” (Institute for Politics, Democracy, and the Internet 2006: 16). There is also an age skew: microdonors “were older, on average, than young donors, but there were also eight times as many young people among small donors as there were in the top group” (Institute for Politics, Democracy, and the Internet 2006: 16). Both large and microdonors tend to be motivated by ideological considerations, but the 2004 study found “little evidence to support the concern that small donors tend to be more polarized and more negative than large donors” (Institute for Politics, Democracy, and the Internet 2006: 27.) We will

The Changing Nature of Campaign Financing for Primaries

41

have to see if these trends hold up in future studies of the much larger pool of microdonors in the 2008 presidential election. In sum, though microdonors are not perfectly representative of the American voting population, they are more representative of the American population than large donors, who have dominated the system in the past. Moreover, the rise of microdonors shows that a candidate with broad appeal (such as Obama or McCain) or a candidate with narrow or intense appeal (such as Paul or Kucinich) can raise enough money from people with modest means to get out the candidate’s message and make the candidate a credible one if there is some popular support for the candidate’s ideas. It is harder for big money to drown out the voices of the many. This is political equality that “levels up” the field rather than “leveling down” with spending limits (Fleishman and McCorkle 1984). Indeed, the amounts of money that a popular candidate is able to raise from microdonors can allow the candidate to compete effectively even against self-funded wealthy candidates. Contrast the size of microdonations raised by Obama and McCain to Mitt Romney’s failed 2008 candidacy. Romney donated $44.6 million to his campaign,17 a bit more than McCain’s microdonations and not even a quarter of Obama’s. The possibility of a wealthy candidate scaring away less wealthy competitors is less credible than in the past.

Candidate Attributes and Quality If we measure progress by the number of microdonations or microdonors, we could conclude that the campaign finance system is moving in a positive direction. But that’s not the only rubric for measurement. Consider also candidate attributes and quality. The donor pool helps to determine which candidates are viable ones, and the changes in campaign financing could affect which candidates are able to compete effectively for the nomination. For example, to be able to attract a very large number of microdonations to run a viable presidential primary campaign, a candidate will have to build a large number of supporters through various media such as television and the Internet. This should give an advantage to candidates with “star power” who can command media attention. Obama’s ability to attract a huge number of microdonations was driven in part by his star persona.

42

Chapter 1

Star candidates may or may not be high-quality candidates, however we measure quality. But it is certainly possible that some high-quality but low-key candidates who might have been able to raise sufficient funds under the old campaign finance system will have a harder time raising funds under the new system. A system dominated by microdonorfunded candidates could drive high-quality but low-key candidates out of the political race. That’s less a danger because the current system and any future system would not rely solely on candidates driven by microdonors. Under the current “invisible primary” system, candidates early in the process get seed money to become viable by making pitches to party activists willing to give maxed-out donations (currently at $2,300). Even a star candidate such as Obama went through the invisible primary in the 2008 primary season. Through the first quarter of 2007, at the beginning of the 2008 nomination process, Obama raised 68 percent of his contributions from donors giving at least $1,000 (46 percent of the total from maxed-out $2,300 donors) and only 22 percent from microdonations18 (Campaign Finance Institute 2007). For good or for bad, the invisible primary will likely live on, even in the age of the microdonor. Microdonations become more important as the election season progresses, and candidates seek additional support through media exposure to gain party nomination and ultimately the office of the presidency. But the rise in microdonations likely won’t have an effect on which candidates make it to the first round of serious contention. These candidates will need to find the maxed-out donors in the invisible primary to get the media’s attention. Microdonations should matter a great deal a bit later in the process, when a primary race gets competitive. We do not know whether Clinton would have won the 2008 Democratic presidential primary if Obama had not been able to rely on the vast pool of microdonations to come his way thanks to the rise of cheap speech, but it is certainly possible.

CONCLUSION The story of money in presidential primaries in the last three election cycles is a story of change. To begin with, presidential primary candi-

The Changing Nature of Campaign Financing for Primaries

43

dates have had a greater demand for money. Since the Bush 2000 campaign refused voluntary public financing with its accompanying spending limit, a candidate arms race began. One of the great tools in that arsenal, pioneered by Bush in an institutionalized way, is the use of bundlers. Along with an increased demand by candidates for money during these last three election cycles, we have seen an increase in supply. Supply increased because of changes in the law, most notably the doubling of the individual contribution limit, and changes in technology, especially the rise in “cheap speech” facilitated through the Internet and, in the 2008 election cycle, cell-phone texting technology. Bundling and increased contribution limits allow candidates to take in great amounts of money in exchange for candidate access. Cheap speech facilitated the rise in microdonations. The Obama fundraising phenomenon in the 2008 election may be unique, and it is risky to draw general conclusions about the ability of future candidates to rely on microdonations for their fundraising. Nonetheless, microdonations are not a key component of the early “invisible primary” stage of presidential campaign politics even for a candidate such as Obama. Microdonations become especially important once a candidate has shown some viability, is tracked by the media, and becomes part of a competitive race. Any normative evaluation of the campaign finance changes in the presidential primary system requires a baseline. Looking at the changes from the point of view of political equality, the rise in microdonations is heartening, by giving more Americans an opportunity to influence which candidates are competitive in the presidential election process. The rise in microdonations serves to offset to some degree the political equality benefits of the matching funds portion of the all but defunct voluntary public financing system. But the system may skew toward the interests of microdonors, whose views and attributes do not perfectly mirror the American population as a whole. The other normative measure is candidate quality and attributes. It might be that changes in campaign fundraising, particularly the rise of the microdonor phenomenon, could tend to favor “star” candidates who can compete in mass media for large numbers of small donations. Even if microdonors tend to favor star candidates, however, candidates

44

Chapter 1

without star power still have other routes to raise money to remain competitive. We should end with a caveat. With the 2008 election just concluded, there is a sense that this is a major if atypical election, and that Obama is an atypical candidate. So it might be that some of the conclusions drawn from the 2008 experience will not apply to future elections. But just as other candidates learned from Bush’s 2000 bundling strategy, future candidates have learned from Obama’s microdonor and cell-phone outreach strategy. Candidates adapt to the campaign finance rules around them, and judging by the sums of money raised in the last three elections, they have adapted quite well.

NOTES 1. This chapter considers only campaign financing for major party presidential candidates during the nomination process. It does not consider (1) minor party candidates for president, (2) general election presidential campaign financing, or (3) congressional campaign financing. Portions of this background material first appeared in Hasen (2008b). 2. For more details on the law described in this introduction, see Lowenstein, Hasen, and Tokaji (2008: 792–93, 889–92). 3. There are also easily circumvented state-by-state limits. 4. These figures do not include additional funds (of up to 20 percent of the spending limit) that may be raised for legal and administrative expenses (socalled “GELAC” funds) (Federal Election Commission 2008). 5. Under that system, the party nominee agrees not to raise any funds for his or her own campaign (though there is some fundraising allowed with the nominee’s party). In exchange, the nominee receives a flat grant from the government. In the 2008 election, Senator McCain, the Republican nominee, accepted the grant of $84.1 million; Senator Obama, the Democratic nominee, opted for private financing (Kuhnhenn 2008). 6. Lowenstein, Hasen, and Tokaji (2008: 890). John Connally declined public funding in 1980, and Steve Forbes declined public financing in 1996. 7. During that same period, Senator Obama raised an addition $41.9 million to be used for the general election campaign (Campaign Finance Institute 2008a: table 2.) 8. The Campaign Finance Institute data show total contributions of over $1.2 billion but include the $41.9 million raised by Senator Obama for the general election campaign.

The Changing Nature of Campaign Financing for Primaries

45

9. Candidates are not legally compelled to disclose bundling activity (though the 2007 lobbying reform law now requires disclosure by lobbyists of bundling activity). The figures below depend on voluntary disclosure by the campaigns. 10. The Campaign Finance Institute data show the percentage as 38 percent, but that figure includes money raised by Obama for the general election. I have recalculated the rates using only primary figures. 11. Microdonations also have played a key role in the fundraising for more ideological but nonleading candidates in the nomination process of both major parties. In 2000, Alan Keyes received 80 percent of his funds and Gary Bauer received 60 percent of his funds from microdonations. In 2004, Dennis Kucinich received 70 percent of his funds from microdonations. In 2008, Tom Tancredo received 81 percent of his funds, Ron Paul 63 percent of his funds, and Dennis Kucinich 75 percent of his funds from microdonations (Malbin 2006: table 11.1; Campaign Finance Institute 2008a: table 2). 12. Howard Dean, an early 2004 frontrunner who eventually dropped out, raised 60 percent of his donations in this manner (Malbin 2006: table 11.1). 13. The Obama campaign announced in August 2008 that it had surpassed 2 million individual donors (Political Intelligence 2008). The vast majority of those donors gave less than $200 to the campaign. Individual information on donors giving less than $200 need not be reported to the FEC. At the end of August 2008, the FEC website had information on only (!) 359,455 donors to the Obama campaign, meaning over 1.6 million donors each gave less than $200. 14. Hillary Clinton, an early frontrunner who dropped out after a long race against Obama, raised 36 percent of her funds in this manner (Campaign Finance Institute 2008a). 15. Again, my figures for overall primary spending differ from the Campaign Finance Institute data because I have subtracted out the general election fundraising by Senator Obama. 16. Obama raised so much more than McCain overall during the primary season that there is not a great absolute difference in the amount each candidate raised through bundling: as noted earlier, McCain raised at least $75.7 million in bundled funds while Obama raised $63.3 million. 17. It came in the form of a loan that was never paid back by the campaign. These data come from the FEC’s database available at www.fec.gov. 18. During the same period, Clinton raised 86 percent of her contributions from donors giving at least $1,000 (68 percent from maxed-out $2,300 donors), and only 9 percent from microdonors. During the same period, McCain raised 74 percent of his donations from donors giving at least $1,000 (34 percent from maxed-out $2,300 donors), and only 19 percent from microdonors (Campaign Finance Institute 2007).

2 MONEY IN 2008: A COLLAPSE OF THE CAMPAIGN FINANCE REGIME? Thomas E. Mann

he decision by Democratic presidential nominee Barack Obama to decline the public grant of $84 million for his general election campaign was widely viewed as, and harshly criticized for, administering the final death blow to the presidential public financing system. That system, rightly considered the crown jewel of public financing in American elections, was set up in 1974 as part of extensive amendments to the Federal Election Campaign Act and has been a central feature of presidential elections ever since. In 2000, George W. Bush became the first successful major party candidate to opt out of the nominating season matching funds section of the law and thereby free himself of the spending limits that are linked to acceptance of the public funds. He raised $94 million before his nomination, well over twice the amount Democratic nominee Al Gore was able to raise and spend with the public financing system. President Bush repeated that practice in his 2004 reelection bid and was followed by Democratic primary candidates Howard Dean and John Kerry. Bush and Kerry each raised around $250 million before their party’s convention, well above the $45 million they would have been limited to had they accepted public matching funds. By 2008, opting into the public finance system was widely viewed as a sign of weakness. Early in 2007, Obama and Hillary Clinton announced their decision to forego public funding in the nominating season, and all of the major Republican candidates followed the same path, although John McCain did so only after the revival of his candidacy in early 2008. Obama raised $456 million through August and McCain $216 million.

T

47

48

Chapter 2

While the nomination component of public financing fell into disrepair over a number of years, the general election grant was accepted by all of the major party candidates between 1976 and 2004. Obama’s decision to reject the public grant for the 2008 general election campaign and his extraordinary success in raising private donations in the fall have set an ominous precedent for the entire program. To many supporters of campaign finance reform, this was only the latest in a series of discouraging developments in the financing of the 2008 elections. The amounts of money raised by the presidential candidates ($1.02 billion for the Democrats, $477 million for the Republicans) shattered all previous records. The money primary in 2007 once again sharply narrowed the field of candidates before the first delegate selection event was held. And the disparity in the general election budgets between Obama and McCain, with the former spending approximately $350 million while the latter lived with his $84 million public grant, raised the prospect of money having an undue influence on the outcome of the election. On another front, the new Supreme Court under Chief Justice Roberts, with Justice Alito replacing Justice O’Connor, has moved in a decidedly deregulatory direction, with one decision potentially undercutting a major provision of the 2002 McCain-Feingold law prohibiting corporate and union treasury funding of electioneering communications. James Bopp Jr. and Bradley Smith are busy mapping and executing a very promising litigation strategy aimed at limiting or reversing earlier decisions—extending from McConnell to Buckley and before—upholding various forms of campaign finance regulation. They will likely have a sympathetic Court to work with for some time. An extended political deadlock between Senate Democrats and Republicans over the nomination of Hans von Spakovsky as commissioner of the Federal Election Commission left the agency with only two members (two short of the majority required for formal action) for several months of the highly contested presidential election cycle. And, as in 2004, outside groups threatened to play a prominent independent role in attacking the presidential candidates. Assessing how these developments have altered the regulatory regime and exploring how they might more dramatically reshape the role of money in elections are essential for rethinking the appropriate

2008—A Collapse of the Campaign Finance Regime?

49

agenda for campaign finance reform. This might presage some cooling of the ideologically charged and tendentious debates about campaign finance that have become so depressingly repetitive and sterile. Most of these debates revolve around the alleged objectives and consequences of the Bipartisan Campaign Reform Act of 2002, widely known as McCain-Feingold. Let me be clear about my own perspective on this matter. McCainFeingold was a very limited legislative initiative designed to restore the effectiveness and credibility of long-standing contribution limits and restrictions on the use of corporate and union treasury funds in federal elections. Its two major pillars—a ban on party soft money and the regulation of electioneering communications—were agnostic toward the total amount of money raised and spent in federal elections even though the rhetoric of some of the bill’s supporters in Congress and outside reformers made clear they longed for a reduction in the money chase. The ban on party soft money has been remarkably successful: parties and elected officials are out of the business of soliciting large unlimited contributions from corporations, unions, and individuals. For the most part, soft-money contributions have not been diverted to outside groups. And the parties have adapted very well to a post–softmoney world. They are today more significant players in the financing of presidential and congressional elections than they were before the enactment of McCain-Feingold. The electioneering communications section of the law was limited to candidate-specific broadcast, cable, and satellite communications to targeted audiences in close proximity to primary and general elections. Its purpose was to extend the existing prohibition on corporate and union expenditures on express advocacy to what had become its functional equivalent. McCain-Feingold left untouched the many avenues for campaigning available to corporations and unions not covered by the bright-line test for electioneering (or by the ban on party soft money). The second pillar of McCain-Feingold also imposed no new restrictions on large individual donors. Preexisting law set a limit of $5,000 for individual contributions to nonparty political committees. The electioneering provision of McCain-Feingold largely achieved the limited goal it set for itself, although that goal increasingly appears circumscribed by outside group activity unaddressed and unaffected by

50

Chapter 2

the law. Moreover, the successful as-applied challenge in FEC v.Wisconsin Right to Life is likely to open new opportunities for candidate-specific advertising supported by corporations and unions. Critics of McCain-Feingold clearly take a much less benign view of the law’s purposes and consequences. Their arguments sometimes emphasize the draconian reach and impact of the law, other times its ineffectiveness and pattern of generating unintended negative consequences. Most prominent among the former is that campaign finance law, McCain-Feingold in particular, suppresses free speech. The argument is decidedly theoretical, not empirical. By virtually every indicator available—ads broadcast, dollars spent, the diversity of views expressed in all forms of campaign activity—political speech has been alive and well in the elections since the enactment of McCain-Feingold. Compared with most other democracies that routinely prohibit paid campaign ads, limit expenditures by parties and candidates, and ban independent expenditures by groups and individuals, free speech in this country is guaranteed by a powerful First Amendment. Repeated charges by such serious analysts as columnists David Broder, Robert Samuelson, and George Will that McCain-Feingold imposes a reign of censorship demonstrates how impervious beliefs systems are to contradictory evidence. It is not surprising that campaign finance critics scrambled to generate a test case to bring an as-applied challenge, since no real ones were available. Attorney James Bopp Jr. searched nationally for groups that would be willing to run an appropriate ad and eventually struck pay dirt in Maine and Wisconsin. This was a perfectly legitimate legal strategy that proved successful but hardly evidence of the suppression of political speech. Critics often argue that the law harms political parties by denying them an important source of income and weakening them relative to unregulated nonparty groups. Yet party fundraising and spending in the 2004 and 2008 elections relative to those preceding McCain-Feingold have been robust, with large sums spent (and more efficiently than softmoney-financed “issue ads”) on behalf of candidates and state and local party organizations. Moreover, the increase in electioneering activities by 527 and nonprofit organizations began before McCain-Feingold was on the books. Clearly, larger forces are at work.

2008—A Collapse of the Campaign Finance Regime?

51

Some have even argued that McCain-Feingold is responsible for the collapse of the presidential public financing system by virtue of its doubling and indexing of individual contribution limits. That is preposterous. Many aspects of the public grant program—state and national spending limits, the size of the match, the timing of payments—fell out of date well before the new law was enacted. George W. Bush successfully opted out of the public match program in 2000 and would have done the same in 2004 without any change in contribution limits. The same is true for Dean and Kerry in 2004 and the major presidential candidates in 2008. Critics also point to the fecklessness of efforts such as McCainFeingold to limit the amount of money in elections or the overall role of wealthy individuals. I agree that such efforts are likely to fail but insist that they were not objectives of this most recent round of reform. The litany of problems associated with money in the 2008 elections conjures up an image of a broken system that is much more pessimistic than the evidence warrants. The large amounts of campaign funds raised were indicative of the extraordinarily high level of public interest in the stakes of the election and the choices offered. Fundraising by the candidates was more an indicator of their electoral appeal than a cause of it. In a political environment hostile to Republicans, it was no surprise that Democratic candidates enjoyed a distinct advantage. Large soft-money contributions to parties from corporations, unions, and wealthy individuals (often arranged through intense pressure from elected and party officials) are no longer a part of the picture. Presidential candidates focused on hard-money contributors, which are limited to $2,300 per donor. During the primaries, several of the candidates, Obama in particular, attracted a huge number of small donors via the Internet. He raised $222 million in contributions of $200 or less through August, fully half of his total funds to that point. (Hillary Clinton and John McCain each raised more than $50 million from small donors, roughly 30 percent of their totals.) To be sure, Obama’s initial fundraising in 2007, like that of the other candidates, depended heavily on large bundled contributions, and almost a third of his nomination season budget came from donations of $1,000 and higher. In preparation for the general election campaign, Obama, like McCain, formed joint fundraising committees with national

52

Chapter 2

and state party committees to take advantage of higher contribution limits to parties than to candidates. Nonetheless, the elimination of large soft-money contributions and the enlistment of millions of small donors indisputably broadened and democratized the contributory class in this election cycle. Money was not the decisive factor in either party’s nomination contest. Although falling short of Obama’s fundraising, Clinton raised enough money to give her a fair shot at winning the nomination. She won a number of primaries in which she was greatly outspent by Obama. Her problems stemmed more from campaign strategy and choices regarding the allocation of her resources. John McCain won the Republican nomination in spite of his fundraising shortfall. His fundraising success came after he became the presumptive nominee. And the efforts of outside groups to shape the election outcome constituted more talk than action. Given the potential legal liabilities and questions about the efficacy of independent advertising campaigns, many wealthy donors decided to stay on the sidelines this election season. Even the seeming collapse of the public financing system has a silver lining. The extended and highly competitive contest between Clinton and Obama, which mobilized millions of new voters and riveted the attention of publics across the globe, would not have been possible if either had opted into the public matching program. Moreover, Obama’s decision to reject the public general election grant did not necessarily alter the balance of resources that would have resulted if both candidates had opted in. That is because the public grant constitutes only a part of the funds spent on behalf of the presidential candidates. In 2008, each major party was allowed to spend $19 million in coordination with its presidential candidate; each party could also follow a precedent set in 2004 and exploit a seeming loophole in the law that allows the cost of “hybrid” ads to be shared by the candidate and his party if they include some generic party references as well as a presidential candidate message. More significantly, the Supreme Court has ruled that political parties may spend unlimited sums on behalf of their candidates as long as they do so independently, while the presidential candidates can assist their parties by transferring unspent funds from their nomination season coffers, sharing donor lists, and engaging in joint fundraising activities. The

2008—A Collapse of the Campaign Finance Regime?

53

McCain campaign budget from the $84 million public grant was supplemented by roughly $72 million in coordinated and independent spending by the Republican National Committee. Obama and the DNC would certainly have exceeded that sum had he chosen to remain within the public funding system. The legal means are available to raise and spend unlimited sums of hard money even for those candidates who accept the public grant. But their success in doing so is shaped importantly by the political environment in which they operate. With a discredited Republican president, a ravaged economy, and an unpopular war, McCain was at a clear disadvantage relative to Obama. The Democratic nominee was certain to have a significant financial advantage with or without public funds. By opting out of the public system, Obama gained strategic control over his resources, allowing his campaign to define the electoral playing field, control his message, and allocate resources in a way consistent with his campaign strategy, and not be dependent on the independent efforts of his party. My generally upbeat reaction to campaign finance developments in the 2008 presidential election and my resistance to the arguments leveled at McCain-Feingold by its critics does not mean that I believe that the traditional approaches to campaign finance regulation will or indeed should have a bright future. The constitutional, political, and practical obstacles to maintaining and strengthening the current regime are daunting. The Roberts Court will look askance at legislative initiatives to shore up the regulatory system as they reconsider previous decisions. New opportunities for fundraising will make it even more difficult to use public funds to entice candidates to limit their spending. Parties and nonparty groups will use their constitutionally protected right to make independent expenditures to prevent any election campaign from being fully financed with public funds and to ensure a cacophony of voices in federal election campaigns. Determined wealthy individuals will avail themselves of myriad legally sanctioned channels for attempting to influence the outcome of elections. Continuing partisan battles over the appointment of FEC commissioners will reinforce the many other sources of its ineffectiveness. Consequently, we are overdue for a reconsideration of campaign finance regulation, one that doesn’t simply rehash the old and stale arguments between reformers and deregulators. Technology is dramatically

54

Chapter 2

reshaping the ways in which funds are raised and campaigns are conducted. The not-too-distant future may witness reductions in the cost of campaigning (with targeted digital communications replacing broadcast ads) and the supplement of a limited number of large contributors with millions of small donors. Candidates and parties will invest heavily in exploiting the potential of the Internet to harness motivated individuals and self-forming social groups, but they may have less control of their campaigns. In this new world, some forms of regulation might stymie rather than facilitate constructive change. At the same time, limits on contributions to candidates and parties may well prove of lasting value. The point is that serious rethinking of the new world of campaigns and campaign finance, one characterized by openness and pragmatism, not by ideological rigidity, is to be welcomed. The 2008 presidential election cycle provided some encouraging developments for those concerned about the role of money in politics. These include the rise of small donors, the failure of some well-funded campaigns and the success of others that were poorly funded, the limitations of paid broadcast ads, the investments in grassroots campaigning, and signs that suggest fundraising is more an indicator than a cause of interest, energy, and electoral appeal. There is no guarantee that these developments will continue into the future and extend to subpresidential elections. The success of small-donor fundraising in the 2008 election cycle will be difficult to match in years without strong national political tides and by candidates without Obama’s star quality. Candidates contesting Senate, House, and state government elections face uncertain prospects in trying to build a base of small donors. If they prove to be ideologically driven, small donors and other Internet-based political organization and communication might drive our politics in an even more polarized direction. Thinking through the rewards and risks of this new world and what public policy steps might encourage a favorable ratio is imperative. A central challenge facing policymakers is to figure out what if anything can be salvaged from the presidential public financing program. Spending limits now drive away serious candidates for each party’s nomination. Raising those limits and making them contingent on all candidates accepting them are almost certain to prove ineffective. Perhaps it is time to return to the idea of public matching funds as seed money for

2008—A Collapse of the Campaign Finance Regime?

55

candidates, released much earlier than permitted by current law and not tied to spending limits. This might provide more lesser-known candidates a better chance of building support for their candidacies during the year of the invisible primary. The general election public grant, originally conceived as a full campaign budget (supplemented by limited coordinated spending by the parties), is now merely one (and oftentimes a modest) component of the total funds spent on behalf of each presidential candidate. Constitutionally protected independent spending by parties makes that inevitable. Under these circumstances, should the general election public grant be retained? And should parties be freed to spend unlimited amounts of hard money (possibly with lower contribution limits than now exist) in coordination with their candidates? Huge challenges lie ahead. Encouraging electoral competition, strengthening transparency, promoting political equality, and reducing conflicts of interest in our electoral system require continuing attention to the role of money. How best to advance those objectives in the new world of technology-driven campaigns should be at the center of the debate on campaign finance reform.

NOTE This chapter is an expansion of an article that appeared in Forum 6, no. 1 (2008).

3 KNOWN UNKNOWNS IN CAMPAIGN FINANCE A l l i s o n R . H ay wa rd

As we know, there are known knowns. There are things we know we know. We also know there are known unknowns. That is to say, we know there are some things we do not know. But there are also unknown unknowns, the ones we don’t know we don’t know. Donald Rumsfeld

hatever other principles divide campaign finance reformers and deregulators, they unite in acknowledging that Congress possesses the power to regulate campaign activity to limit corruption.1 What activity is corrupt, however, is subject to argument. Contribution limits, disclosure of contributions at low itemization thresholds, and other such restrictions address a concern for the corrupt influence of donors on officeholders. Reformers and deregulators part company on the scope, wisdom, or necessity of such restrictions. Whichever side one may be on in this debate, the potential for a donor to corrupt a candidate is a “known known,” to borrow from the analysis of Donald Rumsfeld when he was George W. Bush’s secretary of defense. Editorials and advocates warn us continually about the risks private interests pose to good government. The campaign finance debate is almost entirely about donors and their direct or indirect influence. Not surprisingly, calls for reform arising from the 2008 nomination process concern the relationships donors may have with campaigns and candidates. The prominence of “bundlers” has raised concerns over their indirect evasion of contribution limits, thus providing a means for

W

57

58

Chapter 3

“big money” to buy access (or policy) from recipient nominees. So has the influx of substantial sums from small donors who fall under the $200 itemization disclosure threshold. Expenditures by groups that themselves raise large individual donations have provoked a similar reaction. Hidden beneath the allegations of donor corruption is a different area of manifest corruption, involving the misappropriation of receipts by campaign insiders and the deliberate creation of political organizations for the personal profit of the founders. These are campaign finance’s “known unknowns”—we recognize the potential for malfeasance but have not sought to remedy it specifically through oversight or regulation. It remains untested whether targeted reform could prove successful in combating this other manner of corruption. This chapter argues that the rise of bundlers, small donors, and large-donor independent expenditures should not be greeted with much alarm or calls for greater restrictions. On the contrary, if any campaign finance context deserves additional scrutiny, it should be the internal management of campaigns and committees. In the presidential nomination contest, the fewer publicly financed campaigns subject to mandatory Federal Election Commission (FEC) audits, the greater the risk.2 There should be broad agreement that campaign staff are duty bound to use campaign resources for their intended purpose. Misappropriation is without question corrupt and should be an area where reformers and deregulators can agree that remedies are useful.

BUNDLE OF JOY Bundling has become a topic for attention in light of the success modern presidential campaigns (most notably George W. Bush’s 2000 campaign) have had motivating volunteer fundraisers on their behalf. But bundling in some form has been a campaign fixture for decades. Using well-connected supporters to gather money from friends and contacts is efficient, and necessary for any candidate to broaden his or her circle of supporters. Moreover, existing laws prevent bundlers from coercing contributions from underlings and office mates, or from abusing workplace resources.3 When critics raise objections to bundling, it is usually because bundlers may be able to exercise influence on the candidate, because they can deliver hard-dollar contributions not just from their own re-

Known Unknowns in Campaign Finance

59

sources, but from an entire network of people. Even so, most advocates for additional reform in bundling understand that this networking is necessary, and seek disclosure rather than additional limits.4 Yet advocates of greater disclosure often overlook that disclosure is already required for bundlers who work independently of the campaign.5 It is only those bundlers specially authorized by the campaign to raise funds that need not report separately. Even with them, one could reasonably conclude that the campaign’s own filings detailing all receipts by donor over $200 should be sufficient. Is it necessary for the public (and political rivals and opponents) to know not only who gave what, but who among the campaign’s fundraisers asked? While some argue that additional disclosure would help the voting public understand the interests behind a nominee, there are a number of reasons why more disclosure may be ill advised. First, disclosure chills participation, and we should want to foster bundlers, because they broaden a campaign’s support base. They may also provide an important check on insular campaign groupthink. Moreover, the alternatives to raising money via volunteer bundlers— candidate solicitations and dinners—cost the campaign money and are more vulnerable to incidents of improper influence. True, Internet fundraising has revolutionized small-dollar fundraising, but bundlers reach different donors. A candidate’s Web page cannot substitute for personal contact by these volunteer fundraisers. If bundling becomes an object of greater regulation, we may find that incumbent officeholders do not calibrate the law fairly, and regulation will disadvantage challengers. Whether or not officeholders play political games with the rules, disclosure is more likely to chill bundlers who support third-party candidates, unsuccessful major-party challengers, and primary challengers to incumbents. A bundler for John McCain’s nomination campaign would care less about disclosure than a bundler for the Mike Huckabee, Ron Paul, or Bob Barr committees.6

BIG MONEY IN SMALL PACKAGES Internet fundraising came into its own in the 2008 presidential election. The Obama campaign enjoyed tremendous success in raising contributions through online solicitations, in the prenomination campaign as

60

Chapter 3

well as in the general election. According to a Campaign Finance Institute study, Obama raised $222 million in contributions of $200 or less, 49 percent of the total he raised from the beginning of his campaign up to the end of August 2008. Obama’s small-donor total exceeded the total contributions received by the next two most prolific fundraisers, Hillary Clinton ($216.6 million) and John McCain ($216.4 million) to that date. As the institute noted: “Fueled by Obama’s fundraising, smaller donations accounted for 38% of funds raised by presidential candidates through August 2008, an increase from 34% in 2004 and from 25% in 2000.”7 During the primary, enterprising researchers began to observe that the campaign was reporting contributions from foreign nationals and individuals with clearly made-up names (like “Doodad Pro”).8 The campaign’s apparently inadequate controls became an issue in the general election. Even so, online fundraising has a bright future as a way to enable campaigns to raise substantial funds from small donors with relatively little investment or overhead. There will still be a population of donors that campaigns can reach only by direct mail and telemarketing, but we can expect that as time passes, online fundraising will displace these other expensive forms of small-donor contact.

AN AD OF ONE’S OWN Another 2008 development during the nomination period involved a new twist in the continuing expansion of independent nonparty spending. Spending by issue groups on advocacy has been a part of the campaign landscape since the early 1990s, with tremendous growth in “issue advertising” in 1996 and 2000. To be sure, Congress in enacting the Bipartisan Campaign Reform Act (BCRA, or McCain-Feingold) sought to reduce corporate- and labor-sponsored messages via the electioneering communications restrictions. Groups have fought back in the courts, and the scope of BCRA remains under contention in the 2008 cycle.9 The 2008-season twist has not been grounded in this electioneering communications debate but instead involves reviving another form of permissible group activity endorsed by the Supreme Court in its 1986

Known Unknowns in Campaign Finance

61

Massachusetts Citizens for Life decision.10 Most prominently, the American Issues Project (AIP) sought to take advantage of an exception to the federal corporate expenditure ban by organizing as a “qualified nonprofit” as described in the Court’s decision and the FEC’s regulations. AIP then ran advertisements against Barack Obama as presumptive nominee of the Democratic Party. The qualified nonprofit form of organization would permit AIP to raise funds in unlimited sums from individuals and make express advocacy political communications to the public. Only those funds used for express advocacy need be reported on the group’s independent expenditure reports. Most critically, as a qualified nonprofit, the group need not submit to the limits, prohibitions, and reporting requirements faced by political committees. To be sure, critics have raised questions about AIP’s status as a qualified nonprofit. Chief among these are reservations about the group’s purpose. The Court has in other contexts observed that a group with a major purpose of influencing federal elections must follow the law’s political committee restrictions.11 However, the Supreme Court has not considered a case squarely presenting this issue, and lower courts have applied the “major purpose” standard inconsistently.12 The FEC has never promulgated a definition of “major purpose.” For its part, the Obama campaign argued that AIP’s major donor was guilty of criminal violations of the FECA, which would require demonstrating a knowing and willful violation of the law.13 In such a vague and transitory regulatory area, it is unlikely the Obama campaign could have supported this claim. Should the continued use of bundlers, and the ever expanding alternatives available to rich individuals through independent groups, cause alarm? Provided the FEC adequately enforces the reporting requirements and other requirements already in the law, there seems little reason for concern. Bundlers play an important role in a candidate’s effort to broaden the base and recruit new supporters, and bundlers should be seen as a positive element in large campaigns—like those supporting major-party presidential nominees. But the associational information implicit in volunteer fundraising and networking is sensitive, personal, and worth protecting from intrusive eyes. Greater disclosure would burden bundling, would not give voters sufficient additional information to

62

Chapter 3

justify the burden, and would chill fundraising for also-rans and thirdparty contenders. Responding to the Obama campaign’s small-donor allegations, some reformers have called for greater checks on credit-card contributions, which of course will come at the campaign’s expense. Others suggest that itemized donor disclosure be extended to contributions of under $200, which would vastly increase the size of presidential campaign reports and the amount of data. One might wonder whether disclosure is well-served by increasing the size of the haystack in which the newsworthy or criminal “contribution pin” is hidden. As for an expanding role for qualified nonprofits, here too the law provides ample disclosure of political spending by individuals. If AIP and groups like it must bring their activity within the political committee rules, we can expect their wealthy donors to simply spend the money themselves in the next cycle. Pushing the scope of regulation could penalize some political activists in 2008 but would not alter the options available to their peers in the future.

SOUND MONEY Wherever one stands on the question of donors and their corrupting “undue influence,” everybody should be able to agree on a more constrained definition of corruption, as presented by Robert C. Brooks some years ago. Brooks defined corruption as “the intentional misperformance or neglect of a recognized duty, or the unwarranted exercise of power, with the motive of gaining some advantage more or less directly personal.”14 Present laws could do a much better job of attacking this kind of corruption in campaign finance. This “duty driven” corruption occurs in two related but distinct situations. The first is when a trusted staff person embezzles committee funds secretly. The second is when the committee’s principals form the committee explicitly to obtain money for themselves or their associates, rather than for use in political activity. Given that candidates for presidential nomination are abandoning the public matching fund program, and the mandatory FEC audits that go with it, the potential for undetected misappropriation in nominee committees will increase.

Known Unknowns in Campaign Finance

63

Political committees, parties, and candidates’ campaign committees, especially early in the campaign season, are ad hoc organizations staffed by loyalists, acquaintances, and others who may not necessarily employ professional standards in their work. These entities do not consistently employ good internal business practices. Experience shows they are vulnerable to fraud by staff and misuse of funds. This problem is not new. The landmark case concerning Congress’s jurisdiction to regulate campaigns, Burroughs v. United States,15 arose from allegations that principals with the anti–Al Smith Democratic forces in 1928 diverted funds to their own use.16 Senator Henry Cabot Lodge’s secretary was accused of taking funds meant for the Republican Party.17 While poor fiscal management of political committees has been a persistent problem, it only makes headlines when very egregious conduct is exposed, as in the recent case of National Republican Congressional Committee (NRCC) treasurer Christopher Ward. Ward had served at the NRCC for over 12 years and had become its trusted treasurer, as well as the treasurer for a number of congressional members’ committees. The NRCC required regular audits as a check on the financial management of the committee. Rather than employ outside independent auditors as mandated, Ward forged audits to hide his embezzlement of NRCC and other committees’ funds for his own use. He prepared false audits on his own mock-up of Deloitte & Touche letterhead to present to members. He also submitted these documents to support committee bank-loan applications. News reports suggest that he stole over $810,000 in total.18 Ward’s misconduct is not unique, and in recent years campaign frauds seem to have become more common (or at least better publicized). In 2005, a Democratic Senatorial Campaign Committee (DSCC) fundraiser pled guilty to taking over $360,000 in contributions meant for the committee. He had established an Internet bank account in the name of DS Camp Co. and would deposit “DSCC” checks he collected into that account, rather than forwarding them to the DSCC.19 His scheme was detected only when a donor (who had received no acknowledgment for his donation) contacted the DSCC. The committee reviewed its records and saw that the cancelled check had been deposited in the fraudulent account. Fortunately, the DSCC recovered most of its funds.

64

Chapter 3

Rep. Lloyd Doggett’s campaign bookkeeper would also deposit contributions, but not record them, as they came in. Then she would write herself a check for the same amount, later destroying the cancelled checks.20 She issued 81 checks and stole about $180,000 over five years. The bookkeeper had been a well-respected activist in the Austin area and was a trusted member of the staff. She was able to forge checks, destroy cancelled checks, and falsify reimbursements, not just from the Doggett campaign but from two other state campaign committees. The Doggett campaign discovered the missing funds two years after her departure when attempting to reconcile its account. Wealth attracts fraud, and since Senate campaign accounts are relatively large, Senate committees have seen more than their share of misappropriation. An assistant treasurer in charge of two joint fundraising committees for Senator Elizabeth Dole misappropriated almost $175,000 from March 2002 to May 2003.21 The staffer was a campaign veteran and had worked for the presidential campaigns of John McCain in 2000 and Bob Dole in 1996.22 Senator Joseph Biden’s finance aide used over $412,000 in campaign funds to finance luxury automobiles, dinners, and gifts to his numerous romantic interests. He was caught after the FBI received an anonymous tip.23 On conviction, he was sentenced to three years imprisonment.24 Senator Lindsey Graham’s volunteer bookkeeper, a close campaign confidante who had worked for Graham since his first congressional campaign in 1994, misappropriated a reported $215,000 from November 2002 to April 2007.25 She also used the senator’s personal information to obtain a Visa card.26 In 1993, staff for Senator Paul Tsongas pled guilty to stealing $1 million from his 1992 presidential campaign, which may mark the biggest political fraud case ever uncovered.27 Embezzlers have also taken advantage of House campaign committees. In addition to the recent case involving Doggett, Rep. Lois Capps’s treasurer took about $230,000 while serving on the campaign from 1999 to 2004. She falsely reported checks covering her personal spending as “campaign travel” or misreported the check’s payee so that the FEC reports appeared facially legitimate.28 In 2003, Rep. John Boehner’s finance staffer pled guilty to embezzling $617,000 over almost a decade. He wired money from the campaign account to casinos to fund his gambling. He was sentenced to 30 months in prison and ordered to pay full restitution.29

Known Unknowns in Campaign Finance

65

Campaigns are built on a foundation of close personal relationships, and in some embezzlement cases the committee principals handled the matter privately. Reportedly, Senator Daniel Inouye’s committee arranged for its treasurer, whom Inouye described as a “good friend,” to repay $100,000 taken from the committee’s accounts.30 Similarly, Rep. Anne Northup’s campaign committee arranged to have its former treasurer repay $24,000 taken from the committee, plus $17,000 spent on the audit that detected the theft. Neither of these committees pressed charges against the embezzler. Moreover, Northup’s campaign referred to the former treasurer as a “good employee.”31 Senator Trent Lott’s leadership PAC bookkeeper took $85,000 from the PAC to support a drug habit. His family reimbursed the committee and he entered a rehabilitation program.32 It does not appear that the Lott committee or anyone else pressed charges against the bookkeeper. The relationship between Rep. Bud Shuster and his former chief of staff, Ann Eppard, was by all accounts quite close, and her indictment for, among other things, embezzling $27,500 from Shuster’s campaign came out of a grand jury investigation into illegal gratuities, not charges made by Shuster.33 Some common themes emerge from these incidents. The embezzler is often a trusted political activist, long-term friend, or valued ally. In several of the worst cases, the individual came highly recommended as a political professional, had no criminal record, and would have survived a thorough vetting. For instance, the New York Times noted in its coverage of the NRCC embezzlement that Ward’s “reputation as someone who could balance books and fill out complicated regulatory forms was such that word of [his] skills spread. Eventually he was named treasurer for 83 other Republican committees.”34 Certainly, when a candidate assembles staff for a presidential campaign, there is a limited pool of finance professionals from whom to draw. One could pick a well-recommended professional or stick with a longtime supporter or close personal friend. Neither of these alternatives insulates a candidate from theft, as the history demonstrates. Ties forged through years of campaigning may cause colleagues, or the candidate or chief of staff, to overlook suspicious conduct or clues that, in a less intimate context, would be investigated. One would be challenged to find embezzlers in a business context treated with the understanding extended in the Inouye, Northup, Lott, or Shuster cases. So

66

Chapter 3

one can expect that, left to themselves, political committees may fail to adequately protect campaign funds and ensure they are used for their intended purpose. Even committees with the most robust fiduciary intentions can fall short if they are poorly organized. Treasurers and finance staff can execute these frauds when they are free to act independently of effective supervision. Committees may provide inadequate internal checks from the first, or the embezzler may acquire free rein over time. Campaign finance compliance is nobody’s idea of fun—it is complicated, tedious, and arcane. Over time, other staff members in the chain of command may yield their responsibilities (or leave), and the criminally inclined treasurer might become his or her own supervisor. Recently, the FEC crafted internal “best practices” guidelines for political committees. Committees that adopt these practices, but are nevertheless defrauded by staff, will not face FEC civil penalties.35 Many of the FEC’s practices require committees to employ more staff to oversee one another. In large continuing organizations, this requirement is essential and should be practical, but it may erode as the years go by. In smaller committees (e.g., those supporting a fledgling challenger or a third-party nominee), there is likely to be more vigilance and less opportunity for theft, but these committees also depend on friends and family and have fewer resources to allocate to “best practices” staffing. Employees are expensive. So the FEC’s attempt to improve committee internal practices, by offering one standard applicable to any and all committees, may do a better job curtailing bad conduct in some situations than in others. What more should Congress, or the FEC, do about this problem? One positive step would be to simplify campaign finance. The dizzying array of “joint fundraising committees” and “leadership PACs,” for two examples, is driven by other aspects of the federal law. Were the law more straightforward and easier to follow, it would be more likely that modestly trained staff members could check up on each other. It would be less likely that the finance responsibilities for a complex array of accounts would fall to one select person. With some big enterprises, like committees supporting major presidential nominees, complexity is a way of life. Since these entities could become attractive targets for fraud, additional monitoring for them may

Known Unknowns in Campaign Finance

67

make sense. As noted before, when nominees participate in the public matching funds program, the FEC automatically audits the accounts.36 Now that nominees are less inclined to participate, there may be a need for some other check on their finances. In almost every case mentioned above, a review of the committee’s bank statements would have revealed the hidden fraudulent transactions. In committees raising over a certain annual sum (regardless of the office sought), the laws could require either an independent audit or a reconciliation of the committee’s reports with its bank statements.37 To encourage thoroughness and avoid imposing burdens on political association, these materials could be filed with and reviewed by the commission, but not made public. Alternatively, a committee could win an exemption from the filing and review requirement—provided they obtained an adequate level of insurance. Campaigns that obtain insurance for losses related to employee malfeasance could expect insurers to insist on some controls. Thus, monitoring would be secured by contract, rather than government regulation. Presently, at least one insurer advertises a package tailored to political campaigns.38 Its insurance application asks whether the committee requires bank accounts to be reconciled by someone not authorized to deposit or withdraw funds, whether countersignatures are required for committee checks, and how fund transfers are managed and approved.39 Publicly available disclosure records indicate that presidential candidates purchase insurance from a variety of insurers. A review of 2008 presidential campaign reports indicates that insurance is a significant expense, and that no particular provider has become the standardbearer for this market. Looking only at those expenses that might insure the campaign entity against theft or embezzlement, the Obama campaign’s expenditure reports revealed substantial disbursements over the campaign to Wachovia Insurance Services (over $650,000) and AIG ($49,000). Both companies advertise that they provide the type of liability insurance useful in our context.40 The McCain campaign paid AON Risk Services $275,000 and Hanleigh Insurance $80,000.41 AON advertises a variety of insurance products including enterprise risk insurance. Hanleigh appears dedicated to insuring against risks related to specific individuals failing to deliver on promises through death, disability, or other impediments. The Giuliani campaign and the Thompson campaign also paid

68

Chapter 3

AON for insurance, suggesting that the company may be offering a sensible package for these large committees.42 The Clinton campaign paid D. H. Lloyd & Associates, a Washington, D.C., insurance broker that advertises a variety of business insurance products, about $634,000 over the course of the campaign.43 Presidential nominees outpace other large candidate committees in insurance acquisition. A sample review of down-ticket campaign committees suggests that most do not obtain insurance against the risk of losses and liability from (among other things) embezzlement.44 Therefore, most other committees will not have fiscal discipline imposed from the outside by insurers. This would likely change if committees were presented with the alternative suggested above, of reporting financial reconciliations to federal regulators or securing insurance. Other checks could be added to the regulatory system to help detect fraud. Today, federal law does not require individual donors to disclose their contributions. Donor reporting could provide another way to check the completeness of a committee’s report; if the donor disclosed making the contribution, but the committee’s treasurer never reported receiving it, one would know to go looking for it. The pain and confusion involved in donor reporting would not be worth the effort for small donors but might be for individuals who regularly give more substantial sums (say $30,000 annually).

HIDDEN MONEY One of campaign finance’s great ironies is found in the treatment of expenditures. The first publicity reformers were keenly interested in exposing political expenditures.45 Many of these early activists believed that transparency in political spending would curb corruption—in the form of bribes, payments to voters, and other unsavory uses. Yet over the years, the reform impetus moved toward regulating political contributions and focused on donors rather than recipients.46 Researchers today can search federal campaign contributions statements by donor name, occupation, employer, and zip code but have no way to search expenditure statements by name of recipient, employee, location, or description of expense.

Known Unknowns in Campaign Finance

69

Experience suggests this may be an unwise bias, and that reform of real “corruption” should take a hard look at spending. In the presidential primary context, if a campaign takes public money, any abuse of campaign funds for the personal profit of the campaign’s principals can be detected through the postelection audit.47 For example, the FEC ordered Lyndon LaRouche’s 2000 presidential committee to repay over $222,000 in public funds spent for excessive mark-ups to LaRoucherelated vendors.48 As presidential nominees increasingly abandon the public matching funds system, and the accompanying audits, transparency in expenditures is essential as the primary means to identify improper use of funding. For example, Ron Paul’s precedent-setting campaign for the Republican nomination in 2008 set impressive records for single-day fundraising and Internet receipts. Along the way, the campaign reported spending over $14 million of the $34.5 million raised on administrative expenditures.49 Paul spent roughly 40 percent of the funds he raised on administrative expenses, yet this would not seem out of the ordinary. By comparison, Thompson spent about 37 percent and McCain about 23 percent. Barack Obama spent 33 percent of his funds raised on administrative expenditures, and Hillary Clinton about 47 percent. But news accounts report that many of Paul’s staff and his accounting firm are also family members, which may or may not indicate the funds are being properly used.50 Family members would likely be loyal and devoted to the cause, but how much discipline can (or will) the campaign exercise over their spending decisions? Similarly, PACs related to Rep. Linda Chavez made the news in August 2007.51 Reports disclosed that a network of pro-life, anti-union, and conservative committees and foundations operated by Chavez and members of her family funded salaries and other expenses for them, as well as generating funds for more telemarketing and direct-mail fundraising (and legal fees and fines, see MUR [Matters Under Review] 5616). Yet they spent relatively little on contributions to candidates or independent advocacy—the uses one would expect for a PAC’s funds. To be sure, outreach for new donors can be expensive. But other forms of prospecting, notably online, are coming into broad use and proving effective at a fraction of these costs. As time passes, it will be harder to justify operating a group with such high overhead, unless of

70

Chapter 3

course the plan all along is to raise funds simply to pay salaries and fees to the principals who organized the various committees. Part of the problem in monitoring a committee’s use of funds is inconsistent expenditure reporting. Currently, expenditures are reported at the $200 threshold, then described and coded by reporting committees. Coding may not be consistent, and inaccurate descriptions can slip through the cracks.52 Moreover, it is hard to know if a committee payee is a legitimate vendor or a shell entity set up to defraud. But in any case, today the information reported cannot be easily manipulated. One cannot pull a list of expenditures made to a certain vendor or individual, as one could do if seeking a record of a donor’s contributions. Donor records can be retrieved by name of employer or occupation, but expenses cannot be retrieved by description. Accordingly, detection of any impropriety usually requires some other inside information. The controversy surrounding John Edwards’s PAC payments to a video contractor (who was also a romantic interest of his) erupted not from review of public reports but after the tabloid National Enquirer published the report of their Edwards stakeout.53 Without a tip, the research required to detect abuse is too tedious for most investigators. To prepare a broadcast on leadership PAC spending, staff for the program Marketplace spent six months reviewing FEC reports.54 Even groups with impressive investigative resources find it hard to identify which committees make improper use of their funds. In the 2008 cycle, the Center for Responsive Politics attempted to identify consultants and candidates who earned extraordinary (and potentially unwarranted) returns for their work.55 Yet their lists could only report which candidates spent a large percentage of their funds on direct mail and focused on a select set of firms. High direct-mail spending may be a good use of funds, depending on one’s media market and the competitiveness of one’s race. Such a list is only the first step. The group Citizens for Responsibility and Ethics in Washington has publicized campaign committees that paid the candidate’s family members. Yet, as noted above with the Ron Paul committee, family employees may or may not be an abusive use of campaign funds, depending on the specifics.56 Gifts to pet charities and end-of-year staff bonuses may be legitimate, but maybe not when made with the unfet-

Known Unknowns in Campaign Finance

71

tered discretion of the recipient. If Congress’s reformers are persuaded that certain contexts (e.g., family employment, gifts, travel, or consultant commissions) pose a special dangerous temptation to abuse, then it might make sense to follow California’s lead. Recently the California Fair Political Practices Commission, concerned that candidates abused gift, meal, and travel payment, now requires additional details in expenditure reports.57 In any case, apart from what interesting tidbits may appear on expenditure summaries like those available from opensecrets.org used here, political expenditures remain a tedious and opaque subject for research. Reports of abuse are anecdotal, and any deterrent that public exposure might provide remains sporadic. Developing a systematic and comprehensive way to identify abusive uses of political funds should be a project that reform activists, established ideological groups on all parts of the philosophical spectrum, and members of Congress could all embrace. No well-managed candidate committee benefits from the public cynicism that follows exposure of self-dealing by others. At the very least, Congress should direct (and fund) the FEC to make political spending more transparent by creating a searchable database like that available for donor research. Better this than a system that relies for detection and deterrence on celebrity tabloids.

FOLLOW THE MONEY—ALL THE WAY THROUGH Recent developments in campaign finance indicate that constructive anticorruption reforms should focus on the uses of political funds. Our present disclosure regime, and its vigorous analysis by specialists in the think-tank world, tells us a tremendous amount about where the money comes from. It is time to move the focus to its uses. To reduce the instances of embezzlement, imposition of better internal management practices is a necessity. That should already be a priority for any campaign counsel or manager. Yet given the history here, reliance on campaigns is not enough. Measures that simplify compliance would make a committee’s monitoring task easier, but we may also need to require wealthier committees

72

Chapter 3

(which are most likely to fall victim) to obtain insurance or submit to additional disclosures. In cases where the personal enrichment of the committee’s principals is the unstated policy of the committee, enforcement of consistent descriptive and coding standards, coupled with better electronic access and manipulation of reported information, would help watchdogs, rivals, and donors police committee spending.

NOTES 1. Buckley v.Valeo, 424 U.S. 1, 26 (1976). 2. See the chapter by Hasen in this volume, detailing the 2008 candidates’ abandonment of the primary public funding program. 3. A more detailed version of this argument is in Allison Hayward, “Is That a Bundle in Your Pocket, Or . . . ?” Forum 6, no. 1 (2008). 4. See the chapter by Mann in this volume. 5. 11 C.F.R. 110.6(c). 6. Not coincidentally, while the Obama and McCain campaigns provided voluntary disclosure of bundlers, the Barr campaign did not. See “bundlers” tab at www.opensecrets.org/pres08/index. The Center for Public Integrity studied whether an unsuccessful candidate’s bundlers simply migrate to the presumptive nominee and concluded that in many cases they do not. See “Tracking the Great Bundler Migration: Republican Edition,” www.publicintegrity.org/blog/ entry/581/. 7. Campaign Finance Institute analysis of FEC reports of activity through August 31, 2008, www.cfinst.org/pr/prRelease.aspx?ReleaseID=205. 8. Atlas Shrugs blogger Pamela Geller began reporting suspicious Obama donations in early August 2008, yet it was only in October that the issue received national media attention. Neil Munro, “FEC Rules Leave Loopholes for Online Donation Data,” www.nationaljournal.com/njonline/no_20081024 _9865.php. 9. Wisconsin Right to Life v. FEC, 126 S. Ct. 1016 (2007). 10. FEC v. Massachusetts Citizens for Life, 479 U.S. 238 (1986). 11. Buckley v.Valeo, 424 U.S. 1, 79 (1976); Massachusetts Citizens for Life, 479 U.S. 238, 262 (1986). 12. Akins v. FEC, 101 F.3rd 731 (D.C. Cir. 1996) (en banc), vacated and remanded, 524 U.S. 11 (1998); FEC v. GOPAC, 917 F. Supp. 851 (D.D.C. 1996). 13. Obama, “Conservative Groups Battle through DOJ,” AP, August 26, 2008.

Known Unknowns in Campaign Finance

73

14. Robert C. Brooks, Corruption in American Politics and Life (New York: Dodd, Mead, 1910), 46. 15. 290 U.S. 534 (1934). 16. Louise Overacker, Money in Elections (New York: Macmillan, 1932; (Arno Press reprint 1974), 263–66; Frank Getty, “U.S. Charges Bishop Used Cash Himself,” Washington Post, April 10, 1934, 1. 17. “Seeking Trace of $200,” Washington Post, April 4, 1906, 1. 18. Paul Kane and Ben Pershing, “House Republicans’ Audit Shows Treasurer Stole at Least $800,000,” Washington Post, June 13, 2008, A21; Leslie Wayne, “Awash in Contributions, Campaigns Offer Tempting Targets for Thieves,” New York Times, May 27, 2008, A21. 19. Henri Cauvin, “Political Fundraiser Admits Embezzling,” Washington Post, February 16, 2005, B2. He was sentenced to 24 months in prison. Press Release, June 2, 2005. 20. FEC MUR 5811; Steven Kreytak, “Former Doggett Aide Admits Theft, Gets 30-Day Jail Sentence,” Austin American-Statesman, August 30, 2006. 21. FEC MUR 5610. 22. “Senate Campaigns Facing a Crime Wave?” Roll Call, July 21, 2003. 23. Lisa Getter, “Campaign Catching Hands in the Till,” Los Angeles Times, May 31, 2004, A-1. 24. Paul Kane, “Ex-Biden Aide Faces Three Years for Theft,” Roll Call, March 15, 2005. 25. Robert Behre, “Possible Theft from Graham Campaign under Investigation,” Charleston Post and Courier, October 17, 2007; James Rosen, “Graham Fires Aide Accused of Embezzling $215,000,” Herald, October 17, 2007. 26. David Williams, “Oconee Grand Jury Indicts Former Sen. Graham Aide,” Anderson Independent Mail, March 11, 2008. 27. See Response from Doggett for U.S. Congress, MUR 5811 at 8 (describing series of embezzlement cases). 28. “Capps Says Campaign Official Stole $230,000,” Santa Barbara NewsPress, July 16, 2004. 29. Getter, “Campaign Catching Hands in the Till.” 30. “Senate Campaigns Facing a Crime Wave?” 31. Getter, “Campaign Catching Hands in the Till.” 32. “Lott’s Fund Raising Panel Corrects Reports to FEC,” Houston Chronicle, February 11, 2000, A7. 33. Charles Babcock, “Ex-Aide to Rep. Shuster Indicted,” Washington Post, April 10, 1998, A1.

74

Chapter 3

34. Wayne, “Awash in Contributions.” 35. Statement of Policy, Safe Harbor for Misreporting Due to Embezzlement, 72 Fed. Reg. 16,695 (April 5, 2007). 36. 11 C.F.R. 9038.1. Misappropriated funds would be deemed nonqualified campaign obligations. See “Understanding an Audit Report of a Presidential Primary Campaign,” www.fec.gov/audits/understand_pres_audits.shtml. 37. Presently the FEC only audits political committees “for cause” when they appear to fall short of “substantial compliance” with the law. 11 C.F.R. § 104.16. Presidential campaigns that accept public funds in either the primary or the general election must submit to an audit. See 11 C.F.R § 9007.1 (auditing general election campaigns) and § 9038.1 (auditing recipients of primary matching funds payments). 38. “AIG Introduces Coverage for Political Campaign Committees,” Insurance Journal, September 26, 2007. The package, called LexElect, is available on a surplus lines basis. The minimum premium is $5,000. 39. LexElect Application, available at www.clarkeandsampson.com/Interactive LexElect.pdf. 40. Summary of Obama expenditures available through www.opensecrets .org (August 26, 2008). The Obama campaign also spent substantial amounts on health insurance and appears to have utilized premium financing (i.e., a loan that finances the insurance premium), as there is a reported expenditure to a premium financing company for $127,075 dated April 11, 2007. 41. Summary of McCain expenditures available through www.opensecrets .org (August 26, 2008). 42. Summary of Giuliani and Thompson expenditures available through www.opensecrets.org (August 26, 2008). Giuliani paid AON/Albert G. Ruben Co. (an affiliate of AON) a total of $58,574. The Thompson campaign paid AON Risk Services $4,269. AON trades on the NYSE and is the largest insurance and reinsurance intermediary. 43. Summary of Clinton expenditures available through www.opensecrets .org (August 26, 2008). 44. Among those committees victimized by embezzlement, the committees for Capps and Boehner both appear to have acquired insurance through local agents. Doggett’s apparently has not. Review of expenditures of Capps, Boehner, and Doggett committees available through www.opensecrets.org (August 26, 2008). 45. Perry Belmont, “Publicity of Election Expenditures,” North American Review, December 16, 1905, reproduced in Sen. Doc. No. 89, 59th Cong., 1st Session; Overacker, Money in Elections, 20–50.

Known Unknowns in Campaign Finance

75

46. See Robert F. Bauer, “Not Just a Private Matter: The Purposes of Disclosure in an Expanded Regulatory System,” Election Law Journal 6 (2007): 38 (describing and critiquing donor disclosure). 47. See, e.g., Fulani v. FEC, 147 F.3d 924 (D.C. Cir. 1998) (noting overpayments to vendors and absence of payee endorsements on disbursements); FEC v. Automated Business Services, 888 F. Supp. 539, 542 (S.D.N.Y. 1995) (seeking financial information from Fulani vendors who were also longtime political associates). 48. LaRouche’s Committee for a New Bretton Woods v. FEC, 439 F.3d 733 (2006). 49. Data obtained from www.opensecrets.org. 50. Matthew Mosk, “Ron Paul’s Campaign Is a Family Business, FEC Reports Show,” Washington Post, May 27, 2008, A03. 51. Matthew Mosk, “In Fundraising’s Murky Corners Candidates See Little of Millions Collected by Linda Chavez’s Family,” Washington Post, August 13, 2007, A01; David Fredosso, “PAC Man: Post Attack Prompts Fundraising Freeze,” National Review Online, August 20, 2007; FEC MUR 5616, available at www.fec.gov. 52. For example, the RNC classified coordinated expenditures for clothing for vice-presidential nominee Sarah Palin as “campaign accessories.” Patrick Healey and Michael Luo, “$150,000 Wardrobe for Palin May Alter Tailor-Made Image,” New York Times, October 23, 2008, A1; see also Schedule F for FEC 373483, http://disclosure.nictusa.com/cgi-bin/dcdev/forms/ C00003418/373-483/sf. 53. See T. W. Farnam, “PAC Payments by Edwards Appear Legal,” Wall Street Journal, August 16, 2008, A3; “Senator John Edwards Caught with Mistress and Love Child,” National Enquirer, July 22, 2008. 54. “Oh Waiter! Charge It to My PAC,” Marketplace, July 21, 2008, http:// marketplace.publicradio.org. More detailed information and links to a database are at http://marketplace.publicradio.org/features/pac/. 55. See Luke Rosiak, “For D.C. Fundraising Firm under Scrutiny, the Check Is in the Mail,” July 9, 2008, at www.opensecrets.org (studying BMW Direct); Rosiak, “These Candidates Must Have Been Going Postal,” July 21, 2008, at www.opensecrets.org; see also Frank Phillips, “Firm Gets Big Cut of Campaign Donations: GOP Candidates See Little of Funds Raised; Some Clients, Donors Unaware of Numbers,” Boston Globe, June 29, 2008 (discussing BMW Direct). 56. See “A Family Affair” at www.citizensforethics.org/files/afamilyaffair -senate/FullReport.pdf. 57. California FPPC, Candidate Controlled Committee New Requirements for Reporting Expenditures for Gifts, Meals and Travel (June 2008) (describing regulation 18421.7, effective July 2008).

4 ASSUMPTIONS AND REALITIES OF PRESIDENTIAL PRIMARY FRONT-LOADING Andrew E. Busch

very four years, we have an opportunity to test the assumptions political scientists have fashioned about the way the parties select their presidential nominees. In particular, a number of assumptions about the heavily front-loaded delegate selection calendar were highlighted in 2008. The 2008 nominating races illustrated again how difficult it is to make generalizations about a complex system with so few cases on which to draw. Presidential primary front-loading is the process by which delegate selection contests have been increasingly compressed toward the beginning of the delegate selection calendar. In 1960, only 2 percent of Democratic and Republican primary delegates had been chosen by the end of the fourth week of primaries; in 2008, that figure was 57 percent of Democratic and 52 percent of Republican primary delegates. An associated, though distinct, issue is that front-loading pressures have also pushed the starting gate of the delegate selection system earlier and earlier as Iowa and New Hampshire struggle to avoid being leapfrogged. In 2008, this meant a process that began on January 3 and allocated a majority of primary delegates by the end of the day on February 5, when primaries or caucuses were held in around 20 states, including California, Illinois, New Jersey, and New York. (See table 4.1.) Altogether, a larger percentage of primary delegates were chosen by the end of the fourth week in 2008 than in any previous nomination cycle. In comparison to 2004, the process in 2008 was substantially more front-loaded through the fourth week and was still slightly more frontloaded through the fifth week. However, the process was also more

E

77

78

Chapter 4 Table 4.1. Cumulative Percentage of Primary Delegates Chosen by the End of Selected Weeks after the First Primary

4th 1960 D R 1976 D R 1988 D R 2004 D R 2008 D R

5th

7th

12th

2% 6% 20% 2% 8% 23% New Hampshire primary: March 8

73% 71%

17% 19% 35% 15% 19% 30% New Hampshire primary: February 24

57% 55%

42% 49% 51% 49% 54% 56% New Hampshire primary: February 16

79% 78%

22% 59% 80% 17% 51% 71% New Hampshire primary: January 27

80% 71%

57% 61% 66% 52% 62% 65% New Hampshire primary: January 8

80% 74%

Source: 1960, 1976, and 1988 figures from William G. Mayer and Andrew E. Busch, The Front-loading Problem in Presidential Nominations (Washington, D.C.: Brookings Institution, 2004), 6–7; 2004 and 2008 figures calculated by author.

drawn out in 2008 than in 2004, with significantly more primary delegates chosen after the seventh week of primaries, one fact (among many others) that might help explain the extended Democratic race. A number of factors have combined to drive front-loading. Nominating reforms after 1968 contributed to earlier nominating decisions by more tightly binding delegates to candidates. Likewise, the relatively low individual contribution limits included in the campaign finance reforms in 1974 made it more difficult for candidates to remain in the race if they did not do well in the first caucuses and primaries. The Iowa caucuses and New Hampshire primary drew more and more attention from candidates and the media. When other states noticed the early nominating decisions and the benefits to Iowa and New Hampshire of going early, they wanted to go early as well. And at a certain point, the process be-

Assumptions and Realities of Presidential Primary Front-Loading

79

came self-perpetuating: earlier decisions led to front-loading, which led to even earlier decisions, which led to even more front-loading. A substantial number of scholars (as well as practitioners and journalists) have viewed the front-loading phenomenon with dismay. In their view, front-loading has the effect of limiting the field of contenders, greatly enhancing the importance of the “invisible primary” period before voting begins, and making it likely that a candidate will receive the de facto nomination before most voters have focused on the race and before many states have voted at all. The opportunity for voters to exercise second thoughts about a candidate is much reduced. Many have suggested that front-loading has given an unfair advantage to establishment front-runners, though it may also hold the potential to produce the nomination of an unknown whose meteoric rise could bring victory before the nation has had a real chance to assess him or her (see Loevy 2000; Mayer and Busch 2004: 56–94; Norrander 2000b; Wayne 2000; Rapoport and Stone 2000). On the other hand, some have argued that the increased importance of the invisible primary in the contemporary nominating system has had the salutary effect of restoring the party to a place of prominence in the system, effectively undoing some of the negative effects of the post-1968 reforms (Cohen, Karol, Noel, and Zaller 2008). In any event, over the last decade and a half, a number of assumptions have guided the discussion about how the front-loaded process is likely to play out in practice. Several of them proved faulty in 2008. ASSUMPTIONS ABOUT PRIMARY FRONT-LOADING Assumption 1: The heavily front-loaded system requires candidates to be prepared to pay a substantial entry fee to be able to compete. 2008 Result: Confirmed Assumption 2: Front-loading leads inexorably toward an early thinning of the candidate field, thus severely reducing voter choice in later primaries. 2008 Result: Confirmed Assumption 3: Front-loading leads to an early decision and nonmeaningful participation by later states. 2008 Result: Mixed. R true, D not true. (continued )

80

Chapter 4

Corollary: Two Scenarios: A: Front-runner wins by dint of superior financial and organizational resources. He or she can absorb an early blow and overwhelm opponents. Dole 1996, Bush 2000. B: No clear front-runner or underdog breaks out early and momentum carries him or her quickly to the nomination. Kerry 2004. 2008 Result: Mixed. R scenario B realized, D neither true. Assumption 4: Because of early nominating decisions, delegates in general (and the Democratic superdelegates in particular) are only a formality. 2008 Result: Not true. Assumption 5: Because front-loading pressures have pushed up the start date, and because nominations are likely to be decided early, the general election campaign will be unusually long. 2008 Result: Not true.

Assumption 1: The heavily front-loaded system requires candidates to be prepared to pay a substantial entry fee to be able to compete. The predominant explanation for the entry fee has been that the primaries are so compressed at the beginning of the schedule that an underfunded candidate cannot translate an early success into sufficient fundraising in time for the big rush of primaries. The thinking has been: if you do not have most of the money you need for a nationwide campaign going in, it is too late. Additionally, the growing importance of the invisible primary means that campaigns now both objectively require and are subjectively judged on their organization and fundraising long before the voting starts. In 1996 and 2000, the entry fee was widely judged to have been $20–25 million (Elving 1995; Pooley 1998; Ayres 1999; Brainerd 1999). Now it is considerably higher, perhaps $35–$50 million. Although the rise of speedy Internet fundraising threatened it, this assumption remained essentially undisturbed in the 2008 nomination cycle. Of the five candidates who remained viable on February 5— Barack Obama, Hillary Clinton, John McCain, Mitt Romney, and Mike Huckabee—four had raised at least $37 million by January 1. Only Huckabee had not met a high fundraising bar by January 1, having raised only $8 million. On February 5, he remained mathematically viable but

Assumptions and Realities of Presidential Primary Front-Loading

81

Table 4.2. Funds Raised by January 1, 2008, by Candidates Still Viable on February 5 (in millions)

Democrats Hillary Clinton Barack Obama

Republicans $104.6 $101.4

Mitt Romney John McCain Mike Huckabee

$52.9a $37.0 $9.0

a

Excludes large personal loans. Source: Federal Election Commission, 2007 year-end reports.

was almost certainly not a real threat to McCain, having shown in South Carolina and elsewhere that his support was limited to conservative evangelicals and that he was consequently doomed to bump up against an insurmountable ceiling of support nearly everywhere he went. As always, some candidates met the fundraising bar but failed electorally anyway— most notably Rudolph Giuliani and John Edwards. And Ron Paul raised around $30 million by January 1, allowing him to stay in the race and to continue winning around 10–15 percent of the Republican primary vote all the way to the last contests in June. (See table 4.2.) Assumption 2: Front-loading leads inexorably toward an early thinning of the candidate field, thus severely reducing voter choice in later primaries. The evidence shows an increasing pattern of early withdrawals and even nonentries by candidates who, in an earlier system, would almost certainly have remained viable much longer (Mayer and Busch 2004: 40–46). This assumption, too, remained intact in 2008. The Republican field of presidential contenders started with 12 members, the Democratic field with nine. Four Republicans dropped out in 2007, before voting began (Jim Gilmore, Tommy Thompson, Sam Brownback, and Tom Tancredo); another three dropped out in January (Duncan Hunter, Fred Thompson, and Rudolph Giuliani); one withdrew in early February (Mitt Romney). The GOP field was thus cut down to four by February 7, only a little more than one month after the first vote in Iowa, and two of the four were advocacy candidates (or, less charitably, crank candidates): Alan Keyes and Ron Paul. Among Democrats, one member of the presidential field dropped out in 2007 (Tom Vilsack). Another five fell away in January, including three within one week of Iowa: Joe Biden, Christopher Dodd, and Bill Richardson dropped out by January 10, followed by Dennis Kucinich

82

Chapter 4

and John Edwards near the end of the month. Thus, what was a nineperson field was down to a three-person race by January 30, and one of these was a crank: Mike Gravel. Altogether, the casualties in 2007 and January 2008 totaled 14 of 21 candidates in both parties, including respected former governors of Iowa, Wisconsin, and Virginia; a current governor of New Mexico who was a former cabinet secretary; a former U.S. senator; three current U.S. senators; and arguably the most famous former mayor in America. (See table 4.3.) Assumption 3: Front-loading leads to an early decision and nonmeaningful participation by later states. In 2008, this assumption was borne out on the Republican side, where the race was practically over after February 5. Twenty-six states had not yet voted, including such prizes as Ohio, Texas, and Pennsylvania. The early-decision assumption, however, was obviously incorrect in the Table 4.3. Presidential Candidate Withdrawals, 2007–2008

Democrats Tom Vilsack IOWA CAUCUSES Joe Biden Christopher Dodd Bill Richardson Dennis Kucinich John Edwards Mike Gravel Hillary Clinton

2/23/07 1/3/08 1/4/08 1/4/08 1/10/08 1/24/08 1/30/08 3/25/08 6/7/08

Republicans Jim Gilmore Tommy Thompson Sam Brownback Tom Tancredo IOWA CAUCUSES Duncan Hunter Fred Thompson Rudolph Giuliani Mitt Romney Mike Huckabee Alan Keyes Ron Paul

7/14/07 8/12/07 10/19/07 12/20/07 1/3/08 1/19/08 1/22/08 1/30/08 2/7/08 3/4/08 3/27/08 6/12/08

Assumptions and Realities of Presidential Primary Front-Loading

83

Democratic contest. If Hillary Clinton had lost the Pennsylvania primary and dropped out of the race the next day, April 23, her withdrawal would still have marked the end of the longest active presidential nominating race since 1984, when Walter Mondale did not sew up the delegates that he needed to win the Democratic nomination against Gary Hart until the last day of the primary season. As it was, the Democratic race lasted 152 days from the Iowa caucuses until the nomination was effectively decided—almost seven weeks longer than the 1984 contest. The early-decision assumption has carried with it for some time the corollary that an early decision in a front-loaded system might take place by two routes. Those scenarios deserve further examination. In the first and most likely scenario, the front-runner wins by dint of superior financial and organizational resources. As scholar William G. Mayer has shown, in the modern nominating system, the candidate with the best national poll numbers and the edge in fundraising at the beginning of January usually winds up winning the nomination (Mayer 1996; Mayer 2003). In this scenario, the massive organizational and financial requirements of running in a heavily front-loaded system mean that a front-runner can usually absorb a first blow, lose an early contest or two to a surprising challenger, then come back and overwhelm opponents with his or her superior resources. Bob Dole in 1996 and George W. Bush in 2000 fit this model. Many expected Hillary Clinton to do so in 2008; even her early loss in Iowa conformed to the pattern. In the second scenario, either there is no clear front-runner in the contest and someone breaks out early or an underdog breaks out early, and momentum carries him or her quickly to the nomination (Mayer and Busch 2004: 79–81). John Kerry fit this mold in 2004, when his win in Iowa propelled him to the Democratic nomination against a field that had not produced a clear front-runner. Some thought Obama or (less likely) Huckabee might repeat Kerry’s feat after their wins in Iowa in 2008. The second scenario actually played out in a modified way on the Republican side in 2008. By winning New Hampshire, John McCain took control of a race that, on January 1, 2008, had no clear front-runner. The Granite State in turn carried McCain to a string of victories that assured him the nomination on Super Tuesday (with a bump in the road in Michigan). Indeed, McCain’s course in 2007–2008 was remarkably

84

Chapter 4

parallel to that of Kerry in 2003–2004, except that New Hampshire played the breakout role for McCain that Iowa had played for Kerry. (McCain’s virtual tie for third place in Iowa, and the fact that Iowa was won by Huckabee rather than Romney, left McCain alive until New Hampshire.1) On the other hand, neither scenario proved true for Democrats. Obama did not ride Iowa to a quick sweep, nor did Clinton absorb the blow and go on to overwhelm her opponent in short order. As a result, perhaps the 2008 Democratic contest will lead to the revival of a third scenario that was occasionally touted by analysts when front-loading was first widely noticed but that has been seldom heard of since. That scenario was one in which front-loading could lead to stalemate between the major candidates, on the theory that it is too difficult for any candidate to run successfully everywhere at once (Busch and Ceaser 1996:, 347; Issenberg 2008). (See table 4.4.) Table 4.4.

Year/Party

Length of Active Presidential Nominating Contest, 1984–2008

Days from Iowa

Days from New Hampshire

2008/Da 2008/Rb

152 35

147 30

2004/Dc

44

36

2000/Dd 2000/Re

45 45

38 38

1996/Rf

31

23

1992/Dg

38

30

1988/Dh 1988/Ri

73 50

65 42

1984/Dj

106

98

a

Active race ends when Obama reaches majority of delegates Active race ends when Romney withdraws Active race ends when Edwards withdraws d Active race ends when Bradley withdraws e Active race ends when McCain withdraws f Active race ends when Forbes withdraws g Active race ends when Tsongas withdraws h Active race ends when Gore withdraws i Active race ends when Dole withdraws j Active race ends when Mondale reaches majority of delegates b c

Assumptions and Realities of Presidential Primary Front-Loading

85

Assumption 4: Because of early nominating decisions, delegates in general (and the Democratic superdelegates in particular) are only a formality. In this view, arrived at after a number of consecutive nominating races when it was clearly true, the decisive factor in modern nomination races is winning early primaries, appearing inevitable, and drying up funds and media attention for alternative candidates (Wayne 2001: 150–52). On the Democratic side, the superdelegates—officially, Party Leader/Elected Official delegates (PL/EOs)—were instituted in 1984 as a group of unbound delegates, representing about one-fifth of all Democratic convention delegates, who were free to exercise their seasoned judgment and formally re-inject the official party into the nominating process. Even these party and elected officials have subsequently tended overwhelmingly to merely support and ratify the voters’ preferences. Calculating that position in the calendar mattered more than delegates, a number of states in 2008 held their primaries before the beginning of the parties’ officially designated primary windows, despite the risk that they would be deprived of some or all of their delegates. Michigan and Florida violated the Democratic windows and became causes célèbres when the party announced it would not seat any of their delegates at the national convention. Five states violated the Republican window—Michigan and Florida, along with New Hampshire, Wyoming, and South Carolina—and were deprived of half their delegates. In the view of these states, what good would their delegates be if they were chosen after the nomination was effectively decided? From the Republican perspective, it was not at all clear that the miscreants misjudged. Four of the five states played an important role in the nomination contest. New Hampshire launched John McCain; Michigan kept Mitt Romney alive to fight at a critical juncture; South Carolina established McCain as the clear front-runner and ended Mike Huckabee’s hopes for a breakthrough; the Florida primary drove Rudy Giuliani from the race, becoming a pivotal moment in McCain’s nomination. For Democrats, though, as in many other ways, the old assumptions simply did not hold up. Delegates mattered again, and the nation was treated for the first time in years to the spectacle of news agencies keeping a meaningful daily tab on delegate commitments to each candidate.

86

Chapter 4

Delegate strategies mattered again, including Obama’s successful play for the delegates from caucus states, which Clinton largely did not contest (Shepard 2008). Moreover, because the race was so close, even the superdelegates mattered, as everyone knew by early February. Once Super Tuesday (February 5) ended with neither candidate having delivered a knockout blow, it could be seen by most observers that the superdelegates could be pivotal, and that both camps would have to court each one assiduously (Barone 2008b; Rothenberg 2008). Not since 1984—the first year when Democrats used superdelegates—could this be said. Assumption 5: Because front-loading pressures have pushed up the start date, and because nominations are likely to be decided early, the general election campaign will be unusually long. From 1996 through 2004, the early decisions of the front-loaded system produced a dual result: (1) an unusually early de facto beginning of the general election campaign, and (2) an awkward interregnum between the effective nomination decision and the national conventions, a sort of political no man’s land where everyone knew who the nominees would be but the traditional starting point of the general election campaign had not yet arrived. A number of problems could be identified with an early general election and long interregnum, including a potential distortion of governing and severe voter fatigue (Wayne 2000: 129). Consequently, the widespread assumption that both parties’ races might well be over on February 5 spawned an assumption that the nation would be subjected to an unprecedentedly long general election campaign in 2008, including a very long interregnum. In 2008, because of the duration of the Democratic race, this scenario did not materialize. Although McCain and Obama were trading some fire earlier, the start date of the full-fledged general election campaign—the first date by which both parties had effectively identified their nominees—was the latest since 1984. Barack Obama secured enough delegates to win the nomination on June 3, and Hillary Clinton did not give up the fight for the Democratic nomination until four days later. In 1984, Walter Mondale sewed up the Democratic nomination on June 5. (See table 4.5.)

Assumptions and Realities of Presidential Primary Front-Loading

87

Table 4.5. Start of Interregnum or General Election Campaign (date both parties’ nominees are known) 2008 2004 2000 1996 1992 1988 1984

June 3 March 3 March 9 March 14 March 19 April 22 June 5

This late start to the general election campaign—late, that is, by front-loading standards—was accompanied by the lack of a real interregnum. Instead, Obama and McCain began full-scale combat immediately. Thus, in 2008, the transition from nomination into general election conformed to the previous front-loading pattern neither in respect to an early starting date nor in respect to a long transition period.

TOO MANY GENERALIZATIONS, TOO FEW CASES Altogether, the track record of assumptions about front-loading in 2008 demonstrates the classic problem of trying to draw conclusions from a small n. There are no more than two serious nomination races, and often only one, every four years. Now we have found at least one condition, which may or may not be the only such condition, that can produce an outcome that fundamentally contradicts the common assumptions about front-loading outlined above: a two-person race between roughly equally matched candidates with offsetting strengths and weaknesses and stubbornly committed demographic or regional blocs of supporters. Of course, this sort of a contest is rare. The last time a presidential nominating contest was seen featuring such equally matched candidates was in 1976, when Gerald Ford and Ronald Reagan battled right into the convention. In 1980, in a similar race, Jimmy Carter fought off an extended challenge from Edward Kennedy, though at no time after the onset of the Iranian hostage crisis in November 1979 did

88

Chapter 4

Kennedy pose as serious a threat to Carter as Reagan posed to Ford. No other race since 1980 approached this dynamic until 2008. In this respect, it is interesting to compare the 2008 Democratic and Republican contests with the 1976 contests. Many analysts had assumed that it was the Republicans, with their multicandidate field and no clear front-runner, who would deadlock in 2008 (Dunn 2008). Instead, it was Democrats. The multicandidate field broke down quickly, and voters migrated to their second or third choices; the two-person race was built on the stability of their coalitions from an early point. Not only did the 2008 Democrats mimic the 1976 Republicans, but the 2008 Republicans mimicked the fragmented 1976 Democrats (although with a much compressed primary schedule), counterintuitively reaching a decision sooner than their partisan rivals. In Presidential Primaries and the Dynamics of Public Choice (1988), Larry Bartels examined the 1976 Republican and 1980 Democratic nomination contests and found that they had in common two wellknown contenders with clearly defined and distinct constituencies. Under these conditions, there was minimal potential for momentum and maximum potential for an extended race (Bartels 1988). Bartels’s analysis has clear implications for explaining the 2008 Democratic race and suggests that scholarly assumptions about front-loading and scholarly analysis of momentum may have been whistling past each other for some time. Even so, Bartels’s observations do not match up perfectly with 2008. For one thing, although Hillary Clinton might have been as well known as an incumbent president, Barack Obama did not fit in the category of a well-known and substantial figure within his party prior to the 2008 campaign, certainly not to a degree that matched Ronald Reagan in 1976 or Ted Kennedy in 1980. He was, rather, a little-known “outsider” who gained a considerable amount of fame within a very short period. For another, while Bartels’s theory explains why there was no momentum-driven quick victory by the underdog, it does not explain why the other (and more common) front-loading result did not obtain—a relatively quick victory by the front-runner on the basis not of momentum but of superior reserves. Altogether, the 2008 nominating contests have been a salutary reminder that elections are ultimately contests between unique individuals, a fact that is often set aside for the sake of a political science that can

Assumptions and Realities of Presidential Primary Front-Loading

89

make generalizations. Different elections may be analogous in a number of ways, but no two elections are exactly alike.

HOW WILL THE PARTIES RESPOND? Given the unexpected course of the 2008 nomination process, there are a number of possible responses that might be adopted by the parties. It is conceivable that the national parties will seek to impose some order on the process through a common set of comprehensive reforms. Such a course has been urged by scholars for years, but it also seems the least likely outcome. Although there are many good reasons why the parties should be open to comprehensive changes to the front-loaded system, there is one very large obstacle to such changes: the need for and difficulty of achieving interparty coordination. Whoever wins the presidential election has a strong tendency to be satisfied with the nominating system. In the context of 2008, Democrats may not want a repeat of a long contest, and the GOP may decide that it liked having had its race settled earlier. In short, the incentives and desires of each party toward nominating reform rarely align neatly. Nevertheless, in mid-2008, both parties showed that they were willing to reexamine their rules. The Democratic National Convention established a 35-member Change Commission charged with doing just that, and the Republican National Convention authorized the party to create a similar commission with the authority to change the Republican nominating rules prior to the 2012 contest, a major development in itself. Before, although Democrats could change their rules between conventions, Republicans could only change their rules at their convention, thus adding a significant procedural obstacle to the political obstacles already facing interparty cooperation. That obstacle has now been removed. There are a number of other potential responses by the parties that do not require coordination between the national parties (for an early appraisal of the status of some of these debates, see Halloran 2008): 1. Complacency and inaction. The creation of reform commissions is no guarantee of actual reform. While the 2008 Republican convention

90

Chapter 4

made it easier for the party to reform in between conventions, the party’s Rules Committee rejected on the eve of the convention two competing proposals to establish order on the 2012 primary calendar.2 In the end, the parties might be persuaded that front-loading is not as big a problem as many had thought, leading to complacency. Given the difficulties and uncertain consequences of changing party rules, the parties may be forgiven for such complacency. Moreover, there may be some truth in this analysis. Front-loading may indeed not be as big a problem as many thought, but so far this has been shown only in very rare circumstances. 2. Natural or organic back-loading. Due to the nature of the Democratic campaign, some states may regret their decision to schedule their contests early in the nomination schedule. Political Pulse:The Newsletter of California Politics and Government editorialized about California’s February 5 primary in its February 15, 2008, edition: “We end up getting a lot more campaign visits, which was a big part of the goal. But now we’re looking longingly at the late primary states, who may get the final say on the Dem side” (Political Pulse 2008: 2). Indeed, many early voting states ultimately looked to Pennsylvania or North Carolina with envy. It is conceivable that when the dust settles from the 2008 contest, some states will voluntarily move back their primary dates in 2012. History does not move inexorably in the direction of earlier and earlier nomination decisions, and at least some states may respond to that reality. 3. Added incentives to move states back. Scholars have long argued that an effective way to stretch out the nominating process would be to mandate proportional representation in early primaries and some form of winner take all in later primaries (Rapoport and Stone 2000). Democrats might have become open to such a change, as they are seemingly disillusioned with their nominating rules mandating proportional representation in all primaries and caucuses. Republicans have a long tradition of leaving such allocation decisions to the states and might resist such a policy on federalism grounds. They would have to move in the opposite direction from Democrats, using proportional representation more rather than less often and disallowing early winner-take-all states. 4. Holding the line on windows. The mad rush to the front of the 2008 calendar, in which Americans flirted with a nomination calendar

Assumptions and Realities of Presidential Primary Front-Loading

91

starting in December 2007, presented an opportunity for the parties to prove that they would enforce the windows that dictate the starting point of the delegate selection season. The Republican offenders lost half of their delegates; the Democratic offenders lost all of their delegates and were taken out of the game as significant players until the contest was decided, and Obama, now free to be gracious, relented (first allowing half of the delegates and then all). The parties, then, were partially successful in laying down a marker in 2008. It is an open question whether states will learn a lesson, and whether the parties will build on 2008 to firmly enforce windows next election cycle or shy away from the bitterness engendered in the Michigan and Florida fights. Both parties also headed in the direction of moving back the starting point of their windows. The Republican convention voted to prohibit states other than Iowa, New Hampshire, and South Carolina from scheduling their 2012 contests before the first Tuesday in March (the three exceptions may not vote before the first Tuesday of February). Key Democrats also envisioned a window for most states starting in March (“GOP Adopts Rules” 2008; Balz 2008). 5. Superdelegate and caucus reshuffling. What, if anything, will Democrats do with superdelegates? On one hand, the 2008 contest revealed the difficult dilemma facing the superdelegates whenever there is a close race—either the superdelegates confirm the popular decision and render themselves superfluous, or they contradict the popular decision and render themselves obnoxious, a result that would be especially disturbing in a party that elevates principles of openness and popular participation in nominating affairs. On the other hand, Barack Obama could not have won the Democratic nomination if the superdelegates had flocked to Clinton as was initially expected. Obama’s campaign expressed an expectation that the Democratic Change Commission would look to reduce the number of superdelegates, but by how much will be the question. Democrats also anticipated that the commission would examine the procedures in caucus states, where many Clinton supporters complained about mistreatment (Balz 2008; Ohlemacher 2008). 6. Campaign finance revision. Outside of the realm of the parties directly, there may be an opening for reconsideration of campaign finance rules. The April 9, 2008, edition of the online journal The Forum focused on issues of campaign finance, asking the question “Has the

92

Chapter 4

American Campaign Finance Regime Collapsed?” (Forum 2008). Barack Obama’s decision to opt out of public financing in the general election further fueled these concerns. Perhaps in this environment some rethinking about campaign finance is possible (see the chapters by Hasen, Mann, and Hayward in this volume). One can imagine that a change in delegate allocation rules might work symbiotically with a reform of campaign finance to mitigate some of the recurring problems of front-loading. In particular, a significant increase in the individual campaign limit might reduce premature candidate attrition and allow candidates to lose early primaries but remain competitive later in the race, even if they are not the frontrunners (Alexander 1996).

CONCLUSION Ever since the front-loading of presidential primaries became a serious concern of American electoral analysis, political scientists have tried to identify certain predictable properties of the system. Unfortunately, those predictions or assumptions can only be made on the basis of a very small number of observable nomination contests, and they can only be tested once or twice every four years. In 2008, some key assumptions about front-loading remained undisturbed. Concerns that the front-loaded system requires a very high fundraising entry fee and that many serious and well-qualified candidates are driven from the race before voters have had a chance to really assess them were not allayed. Concerns that early nominating decisions foreclose debate and deprive later-voting states of meaningful participation were vindicated in the Republican contest. However, the Democratic contest of 2008 defied many of the assumptions that had previously reigned in discussions about primary front-loading. There was no early decision; the front-runner did not bounce back from an early setback with the help of superior resources, nor did an underdog ride an early upset to a quick momentum-driven triumph. Individual delegates and delegate selection rules mattered again; it was better to be Pennsylvania voting late with all of its delegates than Florida voting early with none. The general election cam-

Assumptions and Realities of Presidential Primary Front-Loading

93

paign did not fully get underway, with both parties’ putative nominees assured, until June, the latest start since 1984. To be sure, the circumstances that led to the surprising Democratic outcomes were unusual and not likely to be often repeated. And the Obama victory was a close-run thing; Clinton could well have reversed her fortunes with a few better breaks or a handful of different strategic decisions. The assumptions themselves need not be thrown overboard— they did, after all, continue operating on the Republican side, and there is every reason to believe they will continue operating more often than not in the future. Moreover, many other assumptions about the nominating process remained intact in 2008 (see the chapter by Norrander). There continues to be considerable value in attempting to identify wellgrounded generalizations about the nominating system, or for that matter the operation of other institutions in the political world. Indeed, a large part of what political science is about is using systematic observation to formulate such generalizations. But the surprises of 2008 nevertheless should remind us that electoral systems are not like gravity or the second law of thermodynamics. The study of politics may be a science, but it is one that is not susceptible of the degree of precision and certainty sometimes found in the hard sciences. Real elections with real candidates and real voters will continue from time to time to confound the best-reasoned assumptions of political scientists.

NOTES 1. Fred Thompson finished third in the 2008 Iowa Republican caucuses, with 13.37 percent of the vote; McCain finished 273 votes behind, with 13.13 percent. The results were widely reported as a tie for third place. 2. The “Ohio Plan” was similar to the “Delaware Plan” nearly adopted in 2000 and would have required the least populous states to vote first, the most populous states last, in hopes of extending the race; the “Texas Plan” envisioned rotating regional primaries (Ambinder 2008; Hallow 2008; “For 2012” 2008).

5 WHAT POLITICAL SCIENTISTS MAY (OR MAY NOT) KNOW ABOUT PRESIDENTIAL NOMINATIONS B a r b a ra N o r ra n d e r

he 2008 presidential nomination contests were the most expensive in history, excited millions more citizens to vote, and produced the first elongated nomination contest since the 1980s. Conventional wisdom was overturned, and many predictions (e.g., the front-loaded calendar would produce an apparent nominee by February) proved wrong. Does this mean that the existing political science literature is irrelevant? The answer is no. Some models may need tweaking, but the basic components identified by political scientists remained factors in the 2008 contests. This chapter highlights the existing literature and some apparent modifications for the 2008 contests.

T

PRIMARIES AREN’T DIVISIVE One of the criticisms of primaries as they were being introduced at the beginning of the 20th century was that they would produce such fierce intraparty battles that the party would be unable to reunite to win the general election. This only seemed to be confirmed by the fiasco of the 1912 Republican presidential nomination, when Theodore Roosevelt bolted from the party after winning 10 of the 12 presidential primaries but losing the nomination at the convention to President William Taft. Some early political science research supported the notion that primaries were divisive, both at the presidential level and for other offices. These initial analyses compared the divisiveness of the primaries, often measured as a dichotomous classification based on an arbitrary cut-off in 95

96

Chapter 5

the margin of primary victory, with vote totals in the fall election. In research on direct primaries, Hacker (1965) and Pierson and Smith (1975) supported the null effect hypothesis, while Bernstein (1977) found evidence of divisive primaries. The next set of studies (Born 1981; Kenney and Rice 1984) began to include measures of the strength of the Democratic and Republican nominees during the primary (i.e., their overall vote percentage) in comparison with the general election vote. Kenney and Rice (1984) found divisiveness measured in this manner did affect the outcome of primaries for governors and senators, while Born (1981) found no effect for House elections. The third generation of research began to control for weakness among the candidates that would have led to competition in the first place. Thus, Westlye (1991) controlled for the weakness of Senate incumbents and found the effect of divisive primaries to be spurious. Likewise, Kenney (1988) found controlling for candidate quality negated the influence of divisive primaries. Lengle (1980) and Lengle, Owen, and Sonner (1995) translated the aggregate-level methodology to the presidential primaries. Using a 20 percentage-point margin as the definition of divisive primaries, they compared the fall presidential election vote for the parties between states classified as having divisive or not divisive presidential primaries. Kenney and Rice (1987) improved on this model by developing a measure of comparative divisiveness across the two parties’ primaries in a single election and adding some controls for the expected partisan vote in a state, but still found primary divisiveness to harm general election vote totals. Atkeson (1998) faulted prior works for failing to control for other indicators of candidate weakness. She modeled the effect of primary divisiveness (measured in terms of relative primary strength of Democratic versus Republican nominee), unemployment rates, and prior presidents’ evaluations on the vote for the incumbent president’s party in the fall election. When used alone, the relative measure of primary divisiveness is strongly related to the party’s fortunes in the fall election, but when the two controls are added, the divisiveness variable becomes insignificant. Divisive primaries also are investigated from the individual voter perspective to see if those who supported a losing candidate in the primaries are less likely to support the party’s nominee in the fall election.

What Political Scientists May (or May Not) Know

97

Kenney and Rice (1988) examining the 1980 contests did indeed find that those who supported George H. W. Bush and Edward Kennedy were less supportive of their party’s general election candidate (i.e., Reagan and Carter), but in contrast to a depiction of a more divisive primary battle on the Democratic side, the effects were greater on the Republican side. This was because the losing candidate among the Republicans was the more moderate candidate, so that Bush supporters would be more likely to defect to the Democratic candidate than the more liberal supporters of Kennedy would be to defect to Reagan. The relative ideological position of the winning and losing primary candidate may have an effect on the relative harm done to the party’s fall prospects. A number of other researchers have investigated the vote choices of disgruntled primary voters in the fall election. Using the 1980 American National Election Studies (ANES) panel surveys, Stone (1986) also reports an effect for primary divisiveness, but it is an indirect effect through the comparative ratings of the two general election candidates. Southwell (1986), reviewing the 1968 to 1984 presidential elections, found supporters of candidates losing the nominations to be less likely to support the party’s nominee in the fall. Most of these analyses are based on a recall of primary vote preference in the ANES surveys, and Atkeson (1999) finds a significant level of misreporting in these data. Another option for supporters of losing candidates from the primary phase would be to abstain from voting in the fall election. Southwell (1986) finds mixed results. In some election cycles, supporters of losing primary candidates were less likely to vote than other partisans (Kennedy voters in 1968, Muskie and Humphrey voters in 1972, and Brown voters in 1980). In other years, supporters of losing primary candidates were no less likely to vote in the general election (McCarthy and Rockefeller voters in 1968, Wallace voters in 1972, Brown voters in 1976, Kennedy voters in 1980, and Jackson voters in 1984). Finally, in a peculiar pattern (at least from the divisive primary argument), in some elections supporters of losing candidates were more likely to vote than supporters of the winning candidate (Reagan voters in 1976, Hart voters in 1984). Southwell mentions third-party candidacies as offering a choice for disgruntled primary voters that kept them in the fall electorate. Yet the falloff in participation from the primaries to the general

98

Chapter 5

election is minimal. Primary voters are a subset of general election voters, and voters with a strong habit of voting (Norrander 1986a). In addition, the presidential election is not the only contest on the ballot, and those with enough interest to vote in presidential primaries would most likely have interest in participating in choices for governors, senators, or other offices. Divisive primaries also could harm a party’s fall election chances by affecting the role of activists. Those who supported losing candidates in the primaries may be less likely to participate in campaign activities in support of the party’s nominee. Stone (1986), in examining activists from Iowa in 1980, found no carryover effect for these activists’ presidential vote but did find it affected their overall involvement in the presidential campaign. In a subsequent study of caucus attendees from 1984 and 1988, Stone, Atkeson, and Rapoport (1992) found a greater mobilizing than demobilizing effect for the participation of disgruntled activists in the fall campaign. Activists do not appear to be overly dissuaded from participating in the fall campaign if their primary election candidate loses, and a primary race that is mobilizing large groups of new voters, even if some are disappointed in the outcome, should have a larger base of activists for the fall campaign. By the close of the 2008 Democratic National Convention, the divisiveness of the primaries was waning. Forceful statements by both Senator Hillary Clinton and former President Bill Clinton at the convention presented a united Democratic front for the Obama-Biden ticket. Former Clinton supporters continued to move their support to Obama. In the Gallup Poll daily tracker, 81 percent of former Clinton supporters endorsed Obama at the beginning of September, up from 70 percent immediately before the convention (Newport 2008).

GROUP VOTING IN PRESIDENTIAL PRIMARIES: YES, THERE WAS A GENDER GAP BEFORE 2008 Much of the research on voting patterns in the presidential primaries downplays the role of issues. Policy differences between candidates from the same party often are minimal. Partisan voters may care more about a candidate they believe they could trust to make good decisions as a

What Political Scientists May (or May Not) Know

99

president or one who could win the fall election. Thus, candidate qualities often appear to have strong links with voters’ choices, though it is sometimes difficult to tell if this is a true causal relationship. Voters may only be mimicking the campaign themes that they have learned from candidates when they answer survey questions about candidate qualities (Norrander 1986b). A number of groups have had distinctive patterns of support for presidential primary candidates. African Americans have had distinctive voting patterns in a number of the Democratic primaries (Hutchings and Stephens 2008). Of course this was expected in 1984 and 1988 when Jesse Jackson was a candidate, and he won 76 percent of the black vote versus 5 percent of the white vote in 1984, and 91 percent of the black vote versus 11 percent of the white vote in 1988. Al Sharpton in 2004 benefited from 14 percent of the black vote versus 1 percent of the white vote, though Kerry received a majority (56 percent) of the black and white (53 percent) vote. Yet African Americans also favored Carter in 1976 (44 percent of black voters versus 33 percent of white), Clinton in 1992 (71 percent of black voters versus 46 percent of white), and Gore in 2000 (84 percent of black voters versus 68 percent of white). No racial voting gaps appeared in the 1980 contest between Carter and Kennedy. A gender gap appeared sporadically in the presidential primaries in earlier election years (McMullen and Norrander 2000; Norrander 2003). First, a difference exists in the ideological orientation of women and men participating in the primaries. In Democratic primaries, about three-fourths of the time no ideological difference occurs between women and men, but when a gap arises, women are more likely to outnumber men among the liberal category. The gender gap in ideology is actually more pronounced on the Republican side. About two-thirds of the time, women in the Republican primaries view themselves as moderates more often than do men, who are more likely to view themselves as conservatives. Women and men also look for different qualities in potential party nominees. Women look for experience, candidates who care, and candidates who can handle a crisis. Men are more apt to support maverick candidates who can shake things up or, on the other hand, look for candidates who can win the general election. When examining candidate preferences, gender differences in the past also occurred more frequently on the Republican side, with a gender

100

Chapter 5

gap occurring about one-third of the time. Women in 1988 were more likely to support Dole or Robertson over Bush; in 1992 they were more likely to support Bush over Buchanan; in 1996 women supported Dole over most of his competitors; and in 2000 women preferred Bush to McCain. In the Republican Party, women support religious candidates (e.g., Robertson) and avoid protest (e.g., Buchanan) or maverick (e.g. McCain) candidates. On the Democratic side, a gender gap in candidate preferences occurred about one-fourth of the time. In 1980, women supported Kennedy over Carter; in 1984 and 1988 the larger number of African American women voters created a gender gap in favor of Jackson; also in 1984 Glenn and McGovern did poorly among women voters, with Dukakis the beneficiary; and in 1992 Tsongas and Clinton gained more support from women, while Brown did better among men. In 2000, a gender gap occurred in half of the Democratic primaries, with women supporting Gore by 9.6 percentage points more than men. In 2008, a gender gap occurred in almost all of the presidential primaries. Despite media reports that Clinton lost the female vote in Iowa, a gender gap did occur in those caucuses, with women voters being more likely than men voters to support Hillary Clinton. Figure 5.1 shows the gender gap for Senator Clinton in each of the presidential primaries. On average, Clinton received 8.6 percent more votes from women than men, while Obama received 6.9 percent more votes from men than women. Edwards did better among men by 3.4 percent. Clinton had a gender gap among white voters (11.6 percent), Latinos (9.7 percent), and African Americans (4.2 percent). The gender gap for Clinton is a reflection of women wanting to vote for a woman for president, but the campaign themes and images of Clinton versus Obama also fit with prior patterns of gender gaps in desired candidate qualities. Women voters look for candidates with experience and are more reluctant than men to vote for maverick candidates. The Republican Party in 2008 also had some pattern of gender voting. Huckabee on average did better among women by 3.0 percent, reflecting the preference of religiously oriented women within the Republican Party. In contrast, Ron Paul did better among men by 2.7 percentage points. No consistent gender gaps occurred for McCain and Romney; in fact, in some states one candidate would receive more sup-

What Political Scientists May (or May Not) Know

Figure 5.1.

101

Gender Gap in Clinton’s Vote by State.

port from women, but in other states the opposite candidate would have the edge among women voters. A gender gap also occurred in the turnout in the 2008 presidential primaries. Women were more likely to vote in the Democratic primaries, while men were more likely to vote in the Republican primaries. As table 5.1 shows, the gender gap in the 2008 Democratic primaries was Table 5.1. and 2008

2008 Men Women Gap 2000 Men Women Gap

Gender Gap in Turnout in Democratic and Republican Primaries, 2000

Democratic Primaries

Republican Primaries

42% 58% 15% Women

53% 47% 7% Men

43% 57% 14% Women

52% 48% 4% Men

Source: 2000 and 2008 exit polls as posted on the CNN website.

102

Chapter 5

15 percent, but this number was essentially the same as in 2000 when women were 14 percent more likely than men to vote in the Democratic primaries. The gender gap in the Republican primaries in 2008 was 3 percentage points larger than in 2000, with men being more likely than women to vote in the Republican primaries in both years. Most likely this gender gap in turnout in primaries reflects the gender gap in partisanship. Overall turnout in the Democratic primaries increased by 7.7 million voters between 2000 and 2008, based on states with competitive primaries in both years (e.g., a 2000 primary before March 9 when Bradley conceded to Gore), for a 116 percent increase in turnout.1 Turnout in the Republican primaries rose by only 147,259 votes for the competitive primaries (in 2000 before March 9 when McCain withdrew and in 2008 before February 7 when Romney withdrew). The highly competitive 2008 Democratic primaries mobilized a large number of voters. Data posted by Thomas Holbrook on his website suggest that the increase in participation rates was greatest among the young, second greatest for women, and then followed by an increase among African Americans.2 Political science research on turnout in presidential primaries has always demonstrated that a more interesting race (more spending, more competitive contest, more media coverage) draws in more voters (Geer 1989; Moran and Fenster 1982; Norrander 1986a; Norrander and Smith 1985). The 2008 Democratic primary provided the American public with such a contest, and the public responded by flocking to the polls.

CAN WE PREDICT THE WINNERS OF PRIMARIES AND THE CLOSE OF THE NOMINATION CONTEST? A core group of scholars have been forecasting the outcome of the presidential nominations for the 1980 to 2004 cycles (Adkins and Dowdle 2000, 2001a, 2001b, 2004; Mayer 1996, 2004; Steger 2000; Steger, Dowdle, and Adkins 2004). Their models rely on funds raised during the invisible primary stage, cash on hand as of December, and standings in the last Gallup poll before the Iowa caucuses. Some additional fine tuning exists along the lines of adding in results from Iowa and New Hampshire and measures for specific types of candidates. While it may seem that these models are less able to explain the 2008 Democratic

What Political Scientists May (or May Not) Know

103

contest than the Republican contest, the opposite is true. The forecasting models may not have predicted accurately whether Clinton or Obama would win, but they did accurately select these two as the leading candidates and also suggested that the contest would be close. Both candidates raised similar amounts in 2007 (Clinton raised $116 million and had $38 million cash on hand; Obama raised $102 million and had $18.6 million cash on had). In the Gallup poll for December 14–16, 2007, Clinton led with 45 percent of the support to 27 percent for Obama. Clinton and Obama split the Iowa and New Hampshire contests. Thus, two well-financed, competitive candidates produced a long battle for the Democratic nomination. The length of the campaign was longer, but the factors that produced it (money, poll standings, split primary outcomes) were the same. John McCain’s rise to the Republican nomination was more unusual, given past forecasting models. McCain raised $41 million during 2007, lagging behind Romney’s $88 million and Giuliani’s $61 million. The work of Cohen, Karol, Noel, and Zaller (2008), however, would suggest that Romney’s totals needed to be discounted for his own personal contributions. They argue that it is elite endorsements that lead to effective fundraising that can be translated into primary votes. Under this argument, Romney’s 2007 funds raised (from others) totaled $52 million in December, only $11 million more than McCain’s. The December 14–16, 2007, Gallup poll had Giuliani in the lead at 27 percent, with Huckabee, McCain, and Romney in a three-way tie for second (Huckabee at 16 percent, McCain at 14 percent, and Romney at 14 percent). The forecasting models always place more weight on the New Hampshire than Iowa outcomes, so McCain’s victory there helped him more than Romney’s win in Iowa. The other core explanations for the presidential nomination victories are the attrition models (Hanson 2000; Haynes, Gurian, Crespin, and Zorn 2004; Norrander 2006) and end-game predictors (Norrander 2000a). The Republican contest certainly followed this attrition model, winnowing down to John McCain by February 7, when Romney conceded. Yet not to be overlooked is that the Democratic contest also was effectively winnowed early, before becoming a competitive two-candidate contest. Table 5.2 reviews the exit dates of the various candidates. Biden and Dodd dropped out after doing poorly in Iowa, and Richardson left after his poor showing in New Hampshire. The first Republican causal-

104

Chapter 5

Table 5.2.

Attrition of Candidates during the 2008 Nomination Contests

Date

Democratic Candidates

January 4 January 10 January 18 January 22 January 30 February 7 March 4 June 7 June 12

Biden, Dodd Richardson

Edwards

Republican Candidates

Hunter Thompson Giuliani Romney Huckabee

Clinton Paul

ities were Hunter, who was ignored throughout the contest, and Fred Thompson, who did poorly in the first primary in his home region of the South (i.e., South Carolina). Kucinich, who did not leave the 2004 contest until the summer, in 2008 left at the end of January because he faced a tough reelection challenge for his House seat. Edwards dropped out after continuing to finish third, while Giuliani, who had staked his candidacy on Florida, withdrew after failing to win there. Romney dropped out after Super Tuesday. Huckabee and Paul remained in the Republican contest until McCain had secured the official 50 percent of the delegates needed for the nomination. Even Huckabee engaged in speculation that remaining in the contest might hurt his standings in the party, and Paul fits the mode of an advocacy candidate, who stays in the race longer in order to continue publicizing issue positions. The end game for the presidential nominations occurs when the last credible alternative to the eventual nominee leaves the contest (Norrander 2000a). The official number for winning the nominations is onehalf of all delegates. But through candidate attrition, the nomination is often secured before this point. In recent nominations, a margin in delegate strength between the first- and second-place candidates that exceeds 25 percent of the delegates needed for nomination was often enough to convince the last office-seeking candidate to withdraw. Table 5.3 shows how these indicators performed in recent contests.3 In the nine contests with office-seeking candidates (excluding Republicans in 1992 and Democrats in 1996), seven came to a conclusion when the last major opposition candidate exited the race within two or three days of this criterion being met. The two exceptions are the Democrats in 1992, when Tsongas actually left the race too early, and the Democrats

What Political Scientists May (or May Not) Know Table 5.3.

Contest Dem. 1980 Rep. 1980 Dem. 1984 Dem. 1988 Rep. 1988 Dem. 1992

The End Game in Recent Presidential Nominations

Last Office-Seeker Candidate Exit Kennedy None Bush May 26 Hart None Gore April 21 Dole March 29 Tsongas March 19

Rep. 1992 Rep. 1996 Dem. 2000 Rep. 2000 Dem. 2004 Dem. 2008 Rep. 2008

105

Forbes March 14 Bradley March 9 McCain March 9 Edwards March 3 Clinton June 7 Romney February 7

Advocacy Candidate Exit

Jackson None Jackson None Robertson May 16 Brown None Buchanan None Buchanan None

Keyes July Kucinich July 22

Huckabee March 4 Paul June 12

Leader Has 50%

Delegate Margin ⬎25%

June 7

May 10

June 7

April 5

June 9

May 8

April 21 Hart March 24 Jackson April 21 Gore May 5 Jackson March 14 Dole and Robertson March 25 Tsongas March 12 Brown March 12

March 20

March 13

March 14

March 7

March 14

March 7

March 11

March 2

June 3

Never, needed 530 February 5

June 7 April 28 June 4

March 4

Source: Norrander (2000a) and updated by author using CNN delegate counts.

in 2008, where the 25 percent delegate margin was never reached. Neither Obama nor Clinton secured a substantial lead in the delegate totals. Thus, the Democratic contest continued through all of the primary states. On the last primary day, Obama’s total of 2,156 exceeded the 2,118 needed, and Clinton withdrew four days later. In contrast, the 2008 Republican contest, for all practical purposes, was over on Super Tuesday, February 5, when McCain’s delegate count stood at 680 versus 270 for Romney for a margin of 34 percent of the 1,191 needed for the nomination. Romney withdrew two days later. Both the short contest on the Republican side in 2008 and the long contest among the

106

Chapter 5

Democrats came to an end when the delegate totals mirrored those of the end game in previous nomination cycles.

CONCLUSION Political scientists have been studying the postreform presidential nomination system for 32 years and have much to say about the regularities of these contests. The 2008 presidential nominations would, at first glance, appear to be unique battles. On the Democratic side, a long battle ensued between an African American candidate and a female candidate. The contest between these two candidates is far removed from the early 20th century “availability criteria” of white, male Protestants as “ideal” presidential candidates. On the Republican side in 2008, the winner was a candidate with a reputation as a party maverick. Yet, on many fronts, the 2008 contests confirm what political scientists knew from prior research. The 2008 Democratic primary may have been long and produced disgruntlement among the supporters of Senator Clinton at the close of the primary season in June. But by September, the party was not as divided, with 80 percent of Clinton supporters indicating they would vote for Obama. Turnout rose in the 2008 primaries, but for the reasons laid out in previous research: an exciting, competitive contest increases turnout in primaries. Finally, the two contests in 2008 ended as expected. When McCain drew ahead of Romney with a significant delegate lead on Super Tuesday, Romney withdrew. As neither Obama nor Clinton amassed a significant lead in the delegate count, the contests continued through the last round of primaries in early June. Prior political science literature explains much about the regularities in the campaign that led to the unique 2008 outcomes.

NOTES This chapter was originally prepared for the conference “Choosing the President in 2008: The Evolving Process and Its Effects,” held at the University of California, Berkeley, April 10–12, 2008.

What Political Scientists May (or May Not) Know

107

1. The 2008 turnout figures are from the official state websites. The 2000 turnout figures are from Mayer (2001). 2. Thomas Holbrook, University of Wisconsin, Milwaukee, “A Changing Electorate?” http://election08data.blogspot.com (February 21, 2008). 3. Delegate totals for 2000 through 2008 are those reported by CNN on its website. These counts include superdelegates who announced their preferences and estimates of delegates from caucus states.

6 THE RULES AND THE ROLE OF RACE AND GENDER IN 2008 Jennifer A. Steen

he post-1968 reforms were a major step forward in the democratization of the process for nominating presidential candidates, with party elites ceding a major role in the selection of nominees to rank and file voters. The 2008 campaign witnessed a further surge of citizen participation in the decentralized system of state primaries and caucuses. Another step toward democratization in 2008 was the inclusion of serious candidates from what often are described as minority groups. The most compelling story in the election was the spirited contest between a woman (Hillary Clinton) and an African American (Barack Obama) for the Democratic nomination and the ultimate choice of a president who reminded voters that he didn’t look like the icons adorning the national currency. But the 2008 presidential campaign also included other “minority” candidates: a Mormon (Mitt Romney), an evangelical Christian (Mike Huckabee), and a Latino (Bill Richardson) waged credible campaigns, as did a candidate who, if elected, would have been the oldest president ever to take office (John McCain). The 2008 nomination contests thus provided fertile ground for identity politics, a form of voting behavior in which choosing a candidate “like me” is dominant. This chapter reviews the consequences of certain features of the 2008 nominating processes for particular social groups; the principal focus is on race, gender, and Latino ethnicity. Overall, the evidence shows that the leading candidates were not simply “niche” candidates: Clinton, Obama, and McCain each drew significant support from voters who did not share their gender, race, and age, respectively. Yet these same candidates also drew disproportionately from their own identity groups—Clinton consistently

T

109

110

Chapter 6

outpolled all her rivals among female voters, and Obama had overwhelming support from black voters. Similarly, Mike Huckabee’s support was heavily concentrated among evangelical Christians, and Mitt Romney ran well among Mormons. The formal procedures of the Democratic nomination did not inherently advantage Barack Obama as a black candidate or Hillary Clinton as a female candidate. But if one takes a more expansive definition of “process” to include not just the formal rules and institutions in place but the unfolding of events as well, then gender and especially race take on a more significant role. Previous research has demonstrated that, in general, female voters do support women candidates more than do male voters (Dolan 2004; Parkin and Welsh 2008; Sigelman and Sigelman 1982), and black voters are much more likely than whites to support black candidates. These tendencies held in the 2008 contest for the Democratic nomination. However, there were significant variations across state contests in the degree of group voting, and this chapter explores how, if at all, the main features of the nominations process, both formal and informal, amplified or muted the voices of citizens from different social groups. As explained in earlier chapters of this volume, the present system of nominating presidential candidates of the two major parties dates to 1972 when the reforms of the McGovern-Fraser Commission were implemented. The purpose of the new rules, initiated by the Democrats and largely imitated by the Republicans, was to offer “[full, timely and meaningful] opportunities for Democrats to participate in the nominating process” (Polsby 1983: 34). The most significant reform was the general adoption of primary elections and local caucuses open to any party member as the means of allocating delegates pledged to specific presidential candidates. Several changes in the process had potential implications for group voting in 2008, and the next sections of the chapter take them up in this order. Scheduling. The Democratic Party changed its rules for the 2008 election by allowing Nevada and South Carolina to join traditional early birds Iowa and New Hampshire in holding their nominating contests before the designated window for delegate-selection events opened on February 5. Then, disappointed that their bid to join the early states failed, Democratic officials in Michigan defied the national party and unilater-

The Rules and the Role of Race and Gender in 2008

111

ally moved their primary to January 15. In Florida, the Republicancontrolled state government also moved to hold the state primary before the window, on January 29; Florida Democrats seemed torn but were ultimately powerless to stop the maneuver. The Republican Party docked each state half of its delegates, pursuant to the rules of the Republican National Committee, which also treated February 5 as opening day of presidential campaign season. The Democrats’ rules called for both states to lose their delegates to the national nominating convention, although ultimately a milder punishment was administered. Given the significant size of the black and Latino electorates in these states, the effective neutralization of the Michigan and Florida contests in the Democratic race is discussed further below. Front-loading. Within the window for delegate-selection events, the primary calendar also shifted considerably, with states stampeding to the front of the line. On the Democratic side, 22 states (and American Samoa) held primaries or caucuses on February 5, 2008, up from just seven on February 3, 2004. The Republican Party staged delegateselection events in 19 states on opening day, up from only three in 2004.1 By March 4, 2008, there were only 14 Republican and 11 Democratic delegate-selection events remaining; at the same point in 2004 there were 34 Republican and 25 Democratic events still to come. Even the prewindow events moved up: the Iowa caucuses were held two weeks earlier (moving from January 19 to January 3) and the New Hampshire primary moved up three weeks (from January 27 to January 8). Because the demographic composition of state electorates varies, the placement of states’ primaries and caucuses on the calendar may partially determine the relative influence of particular groups on the ultimate selection. Primaries versus Caucuses. The mix of primaries and caucuses has evolved since 1972, with a general trend over time toward more primaries. This trend continued in 2008. The Democrats held 38 primaries and 17 caucuses in 2008 but had held 35 primaries and 20 caucuses in 2004; Republicans held 41 primaries and 14 caucuses in 2008 but had held 32 primaries and 23 caucuses in 2004. Caucuses and primary elections tend to draw different kinds of participants; specifically, lower-income voters are more underrepresented among caucus-goers than among primary voters. Thus changes in the types of nominating events likely affect the overall composition of the pool of nomination participants.

112

Chapter 6

Campaign Finance. In 2008, the limit on campaign donations from individual contributors increased to $2,300 (from $2,000 in 2004), while the overall spending limit imposed on candidates increased to $42.05 million (from $37.31 million in 2004). More critical to the process than these formal changes, however, was the decision by the two leading Democratic candidates to opt out of the public-funding system so that they would not be bound by spending limits for the nominating phase.

NEVADA AND SOUTH CAROLINA JOIN THE EARLY BIRDS Pursuant to a resolution adopted by the 2004 Democratic National Convention, DNC chair Terry McAuliffe appointed members to a Commission on Presidential Nomination Timing and Scheduling. As the commission’s work got under way, representation of minority groups in early-voting states arose as a key issue; however, the establishing resolution barely hints at such concerns. They are completely absent from the eight paragraphs of the preamble, and there is but an oblique allusion in the main motion, an instruction to “[bear] in mind the Party’s commitment to a presidential nominating process that is open and fair to all Democratic candidates and voters” (Democratic National Convention 2004). The commission reported that, in its public meetings, it “heard serious concerns about the disproportionate influence of these traditionally early contests [Iowa and New Hampshire], and the need, early in process, to include states that would be more fully representative of the Party’s rich diversity” (DNC Office of Party Affairs and Delegate Selection 2005: 9). Those concerns were spelled out more bluntly by witnesses at the public meetings, some of whom “agreed that the lack of ethnic and economic diversity in Iowa and New Hampshire prevent minority voters, union members and other core constituencies from having an appropriate influence in the nominating process” (41). The commission also mentioned the front-loading of contests and coordination with the Republican Party as important issues at stake and recommended that the party’s Rules and Bylaws Committee (RBC) take ac-

The Rules and the Role of Race and Gender in 2008

113

tion based on the commission’s recommendations. Upon receiving the commission’s final report, new DNC chair Howard Dean emphasized representation over the other concerns, writing to the leaders of the RBC, “I am pleased that the commission has made recommendations that seek to broaden participation and increase diversity in the early stages of the presidential nominating process” (Lester 2006). The RBC considered proposals from 12 states and the District of Columbia to have their delegate-selection events promoted to the prewindow period. Ultimately, the committee chose Nevada and South Carolina to join the traditional early birds. Nevada, whose 2006 population was 7.3 percent black and 24.4 percent Hispanic, thus leapfrogged from the number 15 spot in 2004 to number four in 2008. South Carolina, with a black population of 28.6 percent, actually dropped two spots due to the overeagerness of Michigan, from number three to number five. However, South Carolina had shared the third spot with six other states in 2004 and in 2008 had exclusive attention for its results. In the words of a prominent RBC member, the shift of Nevada and South Carolina was made explicitly “to bring further regional, racial and economic diversity to the ‘pre-window’ primary period” (Brazile 2007). In other words, given the overwhelmingly white populations of New Hampshire and Iowa, the Democrats wanted to incorporate voices of blacks and Latinos into the early, often determinative stage of the nominating process. (Geography was also explicitly a consideration— Democrats wanted a voice for the South and the West in early contests.)

Potential Consequences: Representation While the promotion of Nevada and South Carolina to an earlier spot on the calendar clearly was made to incorporate the voices of blacks and Latinos, it did not seem to presuppose that those voices would be distinctive. Indeed, party statements, including official documents and public comments by party leaders, treat “diversity” as a goal, not a means to an end. Political scientists label this goal descriptive representation and describe it as the correspondence between the outward characteristics of representatives—usually we think of elected or appointed officials, but in this case the “representatives” are early voters2 with disproportionate influence on the outcome due to their position on the calendar—and

114

Chapter 6

the represented—citizens in later-voting states, whose influence was constrained by the course set in the early-voting states. “In political terms,” Pitkin writes, for descriptive representation “what seems important is less what the legislature does than how it is composed” (1967: 61). Descriptive representation is conceptually distinct from substantive representation, in which the representatives pursue the interests of the represented even if they do not share certain characteristics like race or gender, although substantive representation is often thought to be the primary benefit of descriptive representation. It is assumed that “sharing ascriptive features leads automatically to substantive similarity, that is, sharing perspectives and interests” (Franceschet and Piscopo 2008: 396). If substantive representation of blacks and Latinos in results from early-voting states were a political goal, it would presumably be achieved via descriptive representation, for blacks and Latinos in the population at large by their surrogates in South Carolina and Nevada. Even when shared outward characteristics do not indicate shared political views, descriptive representation may have the advantage of boosting citizen participation. The minority empowerment thesis proposed by some political scientists posits that descriptive representation for underrepresented groups sends an important signal to group members that they can and do have influence on political decisions. Enhanced feelings of trust in government and efficacy in turn motivate civic participation by some group members who would otherwise be bystanders to politics (Bobo and Gilliam 1990). One should thus consider whether black or Latino turnout was bolstered by a schedule shift that was explicitly designed to augment their voices.3 The boost would presumably be greatest in Nevada and South Carolina, but one might also expect blacks and Latinos in later-voting states to be mobilized by their groups’ enhanced clout in the national process. Indeed, caucus turnout in Nevada ballooned from about 9,000 Democrats in 2004 to nearly 120,000 in 2008 (Nevada Democratic Party 2007); however, because of the lack of Nevada entrance polls in 2004, it is unclear whether black or Latino turnout increased more than white turnout in that state. In the South Carolina Democratic primary, turnout surged from about 300,000 voters in 2004 to about 530,000 in 2008, and exit polls indicate that black turnout was especially high. In

The Rules and the Role of Race and Gender in 2008

115

2008, black voters comprised 55 percent of the Democratic primary electorate, up from 47 percent in 2004.4 Unfortunately, disentangling the many determinants of turnout in 2008 is impossible, so the magnitude of the empowerment effect due to the schedule change, if it in fact existed at all, cannot be specified precisely. Both blacks and Latinos were represented (descriptively) in the field of presidential candidates, blacks by Barack Obama and Latinos by New Mexico Governor Bill Richardson, which may have sparked high turnout in both groups.5 The vaunted organizing efforts of Barack Obama’s campaign may also have enhanced black turnout in caucus states, as African Americans were a large and crucial segment of Obama’s supporters. Still, figure 6.1, comparing the Democratic primary electorates in 2004 and 2008, hints that the surge in black turnout was especially pronounced in South Carolina. Among 12 other states for which exit polling data was available for both 2004 and 2008, the average increase in black turnout was 1.4 percent of Democratic voters. In fact, African Americans comprised a smaller share of the Democratic electorate in 2008 than in 2004 in four states. In other words, the relative increase in black turnout evident in South Carolina was exceptionally high.6

Figure 6.1.

Democratic primary electorates in 2004 and 2008.

116

Chapter 6

The turnout statistics from South Carolina provide some support for the notion that African Americans took good advantage of their new position in the calendar, which was intended to enhance the voice of African Americans. But was that louder voice unique in tone? That is, did blacks and Latinos, whose input was channeled through the prewindow Nevada caucus, have distinct preferences such that it would matter substantively if they had more influence on the outcome? As has been widely noted, the answer is yes. Whether one reads state-by-state polls or national surveys, there were obvious, substantial differences among racial groups in terms of their preference for the Democratic nomination. Blacks were overwhelmingly for Obama, Latinos favored Clinton, though by a smaller margin, and whites were split more evenly between the two candidates, although with significant variation depending on gender and education, among other social characteristics. The Nevada results presaged a consistent pattern of divergence between black and Latino voters, with some commentators speculating incorrectly that ethnic competition and prejudice would deter Latinos from voting for Obama in the general election.

Shaping the Course of the Subsequent Campaign The results in South Carolina were distinctive, compared to preceding contests, and as such changed the narrative of the campaign. Subsequent events most likely developed somewhat differently than they would have if attention to South Carolina had been diluted, as it surely would have been had South Carolina shared the spotlight with other states voting on the first day of the window. It does not seem that Nevada’s results placed a distinctive mark on the course of the campaign narrative, as they essentially reinforced the message coming out of New Hampshire and Iowa: this was a very close contest between Obama and Clinton. But when Obama cleaned up in South Carolina, taking 55 percent of the vote (to Clinton’s 27 percent and John Edwards’s 18 percent) and receiving 73 percent of the delegates awarded through the primary, he enjoyed a boost from the positive headlines announcing his victory. The Democratic race through the first contests was characterized as close, but on January 27, large print on the front pages of newspapers told their readers that Obama had tallied an “overwhelming win,” had “trounced” Clinton, and had “[won] big.” Indeed, the headline on com-

The Rules and the Role of Race and Gender in 2008

117

mentary by a prominent political analyst summed it up plainly: “South Carolina Reshapes the Race” (Barone 2008a). In addition, South Carolina seemed to mark a seismic shift in the relationship between race and candidate preference in the Democratic contest. In the lead-in to the first contests, Clinton had won the endorsement of many prominent black leaders and polled well among blacks surveyed. After South Carolina and the public airing of remarks by former President Clinton that were characterized as designed to racialize the Obama campaign, the senator from Illinois became the overwhelming choice of black voters. The results from South Carolina and Nevada (where Clinton had a bigger tally among caucus-goers but the allocation rules favored Obama by one delegate) also influenced the candidates’ subsequent fundraising success. Figure 6.2 depicts the daily difference between the two leading candidates’ itemized contributions from individuals for the first two months of the primary season, along with the average daily difference over three time periods. Obama raised more than Clinton on all but seven of 60 days, but from January 1 through the South Carolina primary the difference between Obama and Clinton was quite small, averaging only about $274,000 per day.7 They were running neck-and-neck in both the delegate-selection events and the money chase. But the day

Figure 6.2.

Individual contributions for the first two months of the primary season.

118

Chapter 6

after Obama’s big win in South Carolina, his daily fundraising advantage spiked to $1.36 million. The day after Super Tuesday, Obama’s advantage rose even higher, to $2.7 million, and his average daily advantage from South Carolina through February 8, the day after Clinton loaned her campaign $5 million, was about $725,000. Clinton’s investment in herself seems to have motivated others to invest in her as well, as she immediately enjoyed two days of a significant fundraising advantage, then stayed fairly close to Obama until the end of February. Indeed, her fundraising exceeded his on four of the last seven days of February after she publicly chastised him for campaign mailings, saying “Shame on you, Barack Obama.” The promotion of Nevada and especially South Carolina to the prewindow period almost certainly affected the campaign, changing the public narrative and the balance of fundraising power. But did either of these consequences ultimately influence the eventual outcome? One can only imagine how the battle would have progressed, and culminated, under a different calendar. But given the extreme closeness of the contest, one can plausibly—although not definitively—argue that Clinton could have staved off Obama’s narrow victory if the states immediately following New Hampshire and Iowa had fewer black voters, or if South Carolina’s results had been buried in Super Tuesday.

THE STAMPEDE TO FEBRUARY 5 Other changes to the primary calendar resulted from states’ own initiatives. The order of state events within the window shifted dramatically in 2008, as more than a dozen states decided to hold their delegateselection events on opening day, February 5. How did the states’ stampede to the front of the line magnify or diminish the influence of black and Latino voters? States with large African American populations generally were positioned to have about the same influence in 2008, with a few exceptions. Obviously, South Carolina (28.4 percent black),8 as already discussed, enjoyed early, exclusive attention for its Democratic primary on January 26. Alabama (26.2 percent black) and Georgia (29.9 percent black) both moved to the first day of the window—Alabama from June 1

The Rules and the Role of Race and Gender in 2008

119

and Georgia from March 2 (Super Tuesday in 2004). However, it is unlikely that either of these moves had much impact on the states’ respective influence over the Democratic nomination. These states sought to achieve more influence by voting early, but in light of how the campaign played out, Alabama and Georgia may have carried more weight had they stayed put with later primaries. On February 5, 2008, 23 states held Democratic delegate-selection events, compared to seven states on February 3, 2004 (the first day of the 2004 window), and 10 that voted on March 10, 2004 (Super Tuesday), causing some observers to label opening day in 2008 “Super Duper Tuesday.” With so many states competing for attention on the same day, the results from any single state received little play outside its borders, with the exceptions of California and, to a much lesser degree, New York. Furthermore, the early primaries were not even decisive collectively, as the battle for the Democratic nomination continued into June. Indeed, Alabama’s results would undoubtedly have received considerably more attention had they come by themselves, or even with one or two others, in early June (when Alabama voted in 2004). Of course, Obama’s victories in Alabama and Georgia contributed to the overall story told about Super Tuesday, which was generally characterized as a push. Obama won more events, but Clinton won in most of the large states, and the delegate allocation from Super Tuesday was virtually even (847 for Obama and 834 for Clinton, a difference of less than 2 percent). Without Alabama and Georgia, Obama still would have won more state contests and Clinton still would have won the contest for delegates by a slim margin (782 to 760). That “what if ” story is simply not markedly different from what actually happened. One consequence of states’ voluntary schedule changes was the discounting of results in Michigan and Florida, both of which have significant black populations (14.1 percent in Michigan, 15.3 percent in Florida). Because these states’ primaries were unsanctioned by national Democratic Party rules, all of the major candidates agreed not to campaign publicly in Michigan and Florida, and four candidates (including Barack Obama and John Edwards, but not Hillary Clinton) removed their names from the Michigan ballot. Although the Democrats’ rules automatically stripped both states of their national convention delegates, a subsequent compromise approved half of a vote for each delegate, allocating

120

Chapter 6

them equally to Obama and Clinton. The Michigan and Florida primary results were thus inconsequential to the delegate count, and because they neither determined delegates nor reflected the level of public support for each candidate, they did not receive the kind of public attention that could have shaped later events. To the extent that the Michigan and Florida primaries signified an opportunity for black voters to influence the nomination, that opportunity was lost. The impact of the jockeying for position on the calendar was similar for Latino voters. States with large Latino populations held about the same spots in the 2008 calendar as in 2004, with two exceptions, Florida and Colorado.9 In Colorado, approximately 12 percent of the electorate is Hispanic. As with blacks in Alabama, this group’s influence on the Democratic nomination was likely diluted rather than enhanced by Colorado’s shift from mid-April to the first day of the window. On February 5, Colorado’s results were lost in the crowd. Had the primary remained mid-April, it would have received more attention. Even if Colorado Latinos lost sway with the calendar change, Latinos as a larger, national group may have had more influence. As noted earlier, 23 states voted on Super Tuesday, and the only individual states to draw individual attention were the two largest, California and New York. Both of these states have significant Latino populations, as do New Jersey, Arizona, Colorado, New Mexico, and Illinois, all of which also held delegate-selection events on February 5. Regardless of the attention paid to these states individually, Latinos contributed significantly to the overall results on Super Tuesday. Indeed, one theme of pre-election headlines advised: “Latino Voters Key to Super Tuesday” (see, e.g., Arnoldy 2008; Roth 2008; Simkins 2008; Tayler 2008). Latinos’ support for Hillary Clinton may well have kept her in the game. Florida’s 3.6 million residents of Hispanic origin constitute more than 8 percent of the U.S. Latino population and 20.1 percent of the Florida population. Florida, of course, had hoped to increase its influence on the presidential nominations by moving its primary up to January 29 (from March 9 in 2004). On the Democratic side, the influence of Florida Latinos was almost certainly reduced by the schedule shift, as described above in the discussion of black influence. The Republican Party’s sanction was less dramatic (although eventually equivalent to the Democrats’), stripping Florida of half of its delegates to the national convention. In the GOP primary, Latinos voted overwhelmingly for the

The Rules and the Role of Race and Gender in 2008

121

eventual nominee, John McCain, who won the Florida primary, and all 57 delegates in this winner-take-all event, with 36 percent of the overall vote. Among Latinos, who constituted 12 percent of the GOP primary electorate, McCain won 54 percent of the vote (FOXNews.com 2008). Latinos, who contributed approximately 6 points to McCain’s total, provided his margin of victory over second-place finisher Mitt Romney. (However, if Latinos had voted for McCain at the same rate as other GOP voters, McCain still would have won narrowly.) His Florida victory put McCain in the front-runner’s spot in the GOP race, although it took him five more weeks to sew up the nomination. In sum, state-initiated schedule changes had a modest effect on the influence of both blacks and Latinos. The overzealousness of Michigan and Florida, which account for 11.4 percent of the black population of the United States, undercut black influence by rendering those states essentially irrelevant. But on average, states with a large number of blacks voted earlier, relative to whites, than they had in 2004. State-initiated schedule changes had a larger impact on Latino influence, making their votes in California and elsewhere critical to Clinton’s survival beyond Super Tuesday.

COMPRESSION OF THE CALENDAR The 2008 election cycle saw unprecedented compression of the presidential nominating calendar, although ironically the Democratic battle continued until the very last primaries were held on June 3. Changes in the schedule meant that after the first day of the window, 32 states had voted (in at least one party), whereas only nine had in 2004. We have already looked at the implications of changes in the order of state events; we must also consider the likely effects of compression qua compression. Generally, front-loading is thought to advantage the front-runner. As Mayer and Busch explain, “With so many caucuses and primaries jammed together, the argument goes, the only kind of candidate who can run effectively—who can campaign in five or ten different states every week—is someone who is already well known and well financed. Someone who is not a first-tier candidate . . . cannot participate in what is, for all practical purposes, a national campaign” (Mayer and Busch

122

Chapter 6

2004: 65). This implies that front-loading disadvantages niche candidates. In an extended campaign season, one can compensate for a fundraising disadvantage with organization; in a compressed season, niche or special-interest candidates do not have time to convert the fervor or organization of their supporters into wider support. As noted above, the 2008 presidential campaign included several “minority” candidates, but because of their appeal to voters from outside their own groups, few observers would describe them as simply “niche” candidates. Compression may have affected how the candidates fared, but it did not affect Barack Obama’s chances just because of his race, Hillary Clinton’s because of her gender, or Mitt Romney’s because of his religion. In fact the closest example of a purely niche candidate in 2008 was a white Protestant male, Mike Huckabee, whose early support came overwhelmingly from evangelicals. For example, exit polls estimated that 83 percent of Huckabee voters in Iowa and 82 percent in South Carolina were born-again Christians. In advance of the actual contests, or perhaps in the very early stages of voting, one would reasonably have predicted that front-loading would disadvantage Huckabee, who had strong evangelical support but not a lot of money. But somehow Huckabee ended up as the little engine that could, picking up steam to become the last competitor standing against John McCain. Huckabee himself remarked, “I’m like one of those tiny little basketball teams in the Final Four, and nobody thought they could get there. I think it’s a credit to the commitment of those who have been supporting me” (Forsyth 2008). Ultimately, compression of the calendar did not affect either the female or African American candidate because in both cases their support went far beyond the confines of their respective groups. The one niche candidate in the race represented a group that is not traditionally considered as underrepresented, and in any case he defied expectations for a narrowly supported candidate.

PRIMARY VERSUS CAUCUS MODE OF DELEGATE SELECTION The nature of the delegate selection process in a state is another factor with the potential to amplify or mute the voices of citizens from various social

The Rules and the Role of Race and Gender in 2008

123

groups. For the Democrats in 2008, 38 jurisdictions held primaries and 17 held caucuses. The Republicans held 41 primaries and 14 caucuses. One can consider how the outcomes in 2008 would compare to systems with all primaries or all caucuses, or to the systems in place in 2004, as 10 states had changed the type of selection event for one or both parties. A caucus is, of course, a much lower-turnout event than a primary. This is illustrated in figure 6.3, which depicts the number of primary or caucus participants as a percentage of the number of general-election votes for the party nominee for the four prewindow states. Like whites, blacks and Hispanics are more likely to participate in a primary than a caucus. But the caucus format most likely filters out more Hispanics and blacks than whites, leaving the minority groups even more unrepresented than in primaries. In the only study comparing primary voters to caucus-goers, Marshall notes, “the low turnout characteristic of caucus and convention systems aggravates fear of serious misrepresentation” and illustrates that “where differences [between caucus attendees and primary voters] proved significant, caucus activists consistently proved of a higher status than those not participating” (1978: 170, 171). Marshall

Figure 6.3. Primary or caucus participants as a percentage of general-election votes in four states.

124

Chapter 6

concludes that “caucuses may appear to perform marginally worse than primaries” (171); however, the difference between primary voters and caucus attendees is much smaller than the difference between either form of participant and nonparticipants. There has been no direct comparison of the racial and ethnic makeup of primary voters to caucus participants, but one can reasonably infer that the status filter of a caucus excludes blacks and Latinos more than whites. If this assumption is correct, the distribution of primaries and caucuses should affect the relative influence of blacks and Hispanics on the outcomes of nominations contests in circumstances where their preferences diverge from each others and those of whites. The implication is that in 2008, each minority group had less influence than they would have if all states used primaries to allocate convention delegates, ceteris paribus, but more influence than if all states used caucuses to allocate delegates. Both groups also should have had more influence in 2008 than in 2004, as four jurisdictions shifted from a caucus to a primary (the District of Columbia, Michigan, North Carolina, and Washington) while only one switched from a primary to a caucus (Nebraska). The jurisdictions that switched to primaries included three with significant black populations, while the one that switched to a caucus did not, so the changes almost certainly improved black representation among participants in Democratic nominating events. And increased participation in North Carolina was especially meaningful, as its results proved decisive (at least in the eyes of über-pundit Tim Russert, who famously proclaimed “we now know who the Democratic nominee is going to be” as the votes from Indiana and North Carolina were tallied on the night of May 6). In contrast, any increase in black participation resulting from Michigan’s switch from caucus to primary was meaningless, as the candidate favored by most blacks was not even on the ballot and the results did not determine any delegates. Whatever influence blacks gained from switches from caucus to primary (through increased participation) was almost certainly undercut by the sanction against Michigan, although this sanction was a result of the schedule, not the type of event. The thought experiment that imagines what would have happened had more states held caucuses instead of primaries must recognize Obama’s remarkable success in caucuses. Ceteris was hardly paribus between caucus and primary states for this reason among others. The or-

The Rules and the Role of Race and Gender in 2008

125

ganizational prowess of Obama’s campaign would likely have compensated for the turnout-depression effect of the more demanding caucus form, although it is unclear that the Obama organizing machine was particularly focused on black supporters. As lamented earlier, the scarcity of caucus entrance polls outside Iowa makes it impossible to establish a baseline against which to measure black (or Latino) participation in the 2008 caucuses. Among whites, Obama was preferred to Clinton by the college-educated and the relatively young, who arguably are more likely to make the extra effort to attend a caucus. So even though black participation in caucuses may have been less, in relative terms, than in primary elections, Obama’s broad appeal and superior organization render this politically inconsequential.

CAMPAIGN FINANCE AND PRESIDENTIAL NOMINATIONS The main formal change to the presidential campaign finance system in 2008 was an increase to $2,300 (from $2,000 in 2004) in the limit on campaign donations from individual contributors. There was no similar increase in the maximum amount of matched contributions from public funds, thus the relative value of the maximum contribution essentially increased. In other words, candidates’ incentive to court high-dollar, “max-out” donors was strengthened, relative to small donors. But the new rule was rendered irrelevant when both Hillary Clinton and Barack Obama opted out of the public-funding system in order to avoid its limits on their total spending. By June 30, 2008, Senator Clinton had raised $213 million and Senator Obama had raised $326 million, both dwarfing the total primary spending limit of $42 million.10 While both of these candidates raised an impressive amount in small contributions during the same period ($165 million for Obama and $66.5 million for Clinton, in unitemized contributions under $200), there can be absolutely no doubt that small donors would have provided a substantially larger proportion of their campaign funds if Obama and Clinton had accepted the matching funds and associated spending limits. Opting out undermines the relative value of small contributions. With matching funds from the federal government, a $250 contribution

126

Chapter 6

is worth $500; without matching funds, it is worth its face value only. All else being equal, one would expect nonparticipating candidates to shift more attention and effort to raising larger contributions. Such a shift could potentially have affected the gender and racial makeup of the contributor pool. If high-dollar contributors are more likely to be whites and males than low-dollar contributors, women and minorities would have played a smaller role as funders of the presidential candidates. There is, to date, no analysis of the 2008 donor pool that reveals whether high-dollar contributors to Obama and Clinton were indeed “whiter” or “maler” than low-dollar contributors. The scant research on other donor pools certainly does not indicate significant differences in the racial makeup of small contributors compared to large contributors, and the indications about the role of gender are mixed. A study of donors to the 2004 presidential candidates suggested no difference in the proportion of women among large and small donors, but it cautioned that “our sample of small donors is not completely representative of all small donors in 2004” (Graf et al. 2006: 13).11 Other work has found that the proportion of women in the donor pool shrinks as contribution levels increase—there are relatively fewer women among large donors than there are among small donors. Joe, Malbin, Wilcox, Brusoe, and Pimlott (2008) examined contributors in state elections and found that women constituted 28 percent of “large” donors ($500 or more), 39 percent of “medium” donors ($100–$499.99), and 45 percent of “small” donors (less than $100). The actual impact of Obama’s and Clinton’s decisions to forego matching funds is impossible to determine, and whatever that impact may have been, it was surely overwhelmed by other factors. While women may usually be underrepresented among large donors, in this cycle many were especially motivated to donate to the female candidate, and Hillary Clinton raised half of her itemized funds (i.e., contributions of $200 or more) from female donors (Schouten 2008). Furthermore, even though opting out devalued donations under $250 by taking away the federal match, both Clinton and Obama raised impressive amounts in unitemized donations. An aspect of the fundraising process in 2008 that did have a disparate impact on the role of men and women in the campaigns is the

The Rules and the Role of Race and Gender in 2008

127

influence of “bundlers.” Bundlers are individuals who solicit contributions for their preferred candidate, then collect and deliver those contributions (or cause them to be delivered) to the candidate. Bundlers are not new to presidential elections and were used to famous effect by George W. Bush in 2000. But they have become increasingly significant over time, and in 2008 both Hillary Clinton and Barack Obama had large networks of volunteer fundraisers who directed huge numbers of individual checks (or credit-card charges) to the campaign treasuries. One can conceive of bundlers in terms of their location in Richard Fenno’s (1978) classic bulls-eye model of a politician’s constituency. Bundlers are a part of the personal constituency, perhaps not as close to candidates as the innermost circle of advisors and key endorsers in earlyvoting states, but not very far from the core. This gives bundlers more access to and more influence with candidates than citizens located on the outer rings. And just as men dominate the ranks of large donors, they dominate the ranks of bundlers. Approximately 68 percent of Hillary Clinton’s bundlers were male, as were 73 percent of Obama’s bundlers.12 So the importance of bundlers is a new aspect of the nominating process, and to the extent that bundling buys influence down the road, the benefits of this role are enjoyed disproportionately by men.

THE RACE AND GENDER GAPS IN VOTER PREFERENCES The formal procedures in neither party directly advantaged (or disadvantaged) Barack Obama as a black candidate, Hillary Clinton as a female candidate, Bill Richardson as an Hispanic candidate, Mitt Romney as a Mormon candidate, or John McCain as a geezer candidate (as he dubbed himself ). Democratic Party rules do specify that half of the national convention delegates must be women; they also require each state party to adopt an affirmative action plan to facilitate inclusion of African Americans, Hispanics, Native Americans, Asian/Pacific Americans, and women. However, those rules do not affect allocation of delegates to specific candidates. In other words, Hillary Rodham Clinton, for example, did not get a leg up in the delegate count because of affirmative action for women in delegate selection.

128

Chapter 6

The familiar gender and racial gaps in voting were manifest in 2008, with women more likely than men to support Clinton and blacks more likely than whites or Latinos to support Obama. It has been less widely noted that there were significant variations in these group differences across individual contests. Here we will consider whether and how these variations related to aspects of the nominating process. The term “gender gap” is used to refer to the difference between the proportion of women favoring a particular candidate (or party) and the proportion of men favoring that candidate. In terms of support for Hillary Clinton, the gender gap among white voters only—Clinton’s share of all women’s votes minus Clinton’s share of all men’s votes— ranged from a low of 2 points (in Indiana) to a high of 21 points (in California). The racial gap in terms of support for Barack Obama— Obama’s share of all blacks’ votes minutes Obama’s share of all whites’ votes—was both larger and more varied than the gender gap for Clinton, ranging from 24 points in New York and New Jersey (Clinton’s home state and its next-door neighbor) to 67 points (in Kentucky). Black women, who may have felt cross-pressured, preferred Obama over Clinton by as little as 13 points (in New York and New Jersey) but by as much as 84 points (in both Illinois and North Carolina). Let us note some discernible patterns in racial and gender voting over the course of the nomination battle, depicted in figure 6.4. First, the gender gap for Clinton narrowed over time among both black and white Democratic voters. Second, the racial gap for Obama widened over time among both men and women, but more markedly among women. Among whites, the narrowing of the gender gap seems to derive from the departure of John Edwards from the contest, which left many white men, and fewer white women, in search of a new candidate. Once Edwards dropped out of contention, Clinton gained support from both white men and women, but she picked up an average of 19 points from white men and 16 points from white women. When the field narrowed to Obama and Clinton, white men and white women both shifted slightly toward Clinton as the campaign wore on (and on), but the correlation between event dates and Clinton’s vote share is small and not statistically significant for either group. Among blacks, Edwards’s departure had no effect on the gender gap. This is unsurprising since

Figure 6.4.

Patterns in racial and gender voting during the 2008 Democratic primary.

130

Chapter 6

Edwards had negligible support from black voters before he abandoned his candidacy. Both black men and black women shifted support from Clinton to Obama over time, and the shift among black women was more pronounced. This resulted in a narrowed gender gap for Clinton among blacks, as well as an increased race gap for Obama among both men and women, but especially for women. Here one must use an expansive definition of “process,” including not just the rules, procedures, and institutions in place but the unfolding of events as well, as it does not seem that the apparent shift over time in black preferences, and stability in white preferences, relates to the nominating procedures themselves. The literature on group voting offers a number of explanations why women and minorities at times prefer candidates from their own groups, but none of them can be connected, even indirectly, to the formal nomination rules, especially so as to explain the temporal variations apparent in 2008. Those explanations generally assign two motivating functions to a candidate’s social characteristics for members of the group that shares such traits. First, certain groups hold distinctive policy positions or ideological orientations, so voters tend to prefer candidates from their own group as a consequence of congruent political preferences or on the assumption that descriptive representation will deliver substantive representation (see, e.g., Paolino 1995). But it is difficult to imagine how this theory would predict increasing racial polarization over the course of a presidential campaign, in which the candidates and their positions are well known at the outset and voter information increases over time. If anything, one would expect voters to rely less on the shortcut of group cues as the campaign progresses and, in the present case, to learn that the policy positions of Obama and Clinton on racial issues were very similar. Second, candidates can benefit politically from a strong sense of group pride among voters from their own groups (see, e.g., Brewer 2001; Huddy 2003; Koch 1994), suggesting that the success of a candidate “like me” or a threat or insult to her (or him) can enhance voting on the basis of a shared background. The tides of change in group voting during the 2008 presidential nomination, particularly in the case of Obama, are largely consistent with this notion. One can plausibly argue that developments in the campaign over time increasingly emphasized group identity for blacks, while gender identity cues remained constant.

The Rules and the Role of Race and Gender in 2008

131

From the outset, Barack Obama presented himself as a “postracial” candidate for all Americans. Initially, he avoided talking about racial issues and the need to make a special effort on behalf of African Americans. Perhaps his most prominent reference to race prior to March 18, 2008, came during his “coming out” speech at the 2004 Democratic National Convention: “There’s not a black America and white America and Latino America and Asian America; there’s the United States of America.” This theme stands in sharp contrast to the presidential campaign rhetoric of Jesse Jackson, who told the 1984 convention, “Our time has come!” and who rallied crowds with “Hands that picked cotton will pick the president!” Obama’s personal biography also lacked strong symbols of African American identity, such as ties to the civil rights movement or even an African American parent (his father was black but a Kenyan, not an American). In late 2007, historian Shelby Steele eloquently described how Hillary Clinton was symbolically “blacker” than Obama, and noted that “much of the excitement that surrounds him comes from the perception that he is only lightly tethered to race” (Steele 2007).13 But other actors brought race to the fore. Of course, the media emphasized the importance of the black vote in the South Carolina primary (January 26). Bill Clinton made a series of remarks that put race into headlines, op-ed pages, talk shows, and blogs, most notably his comparison of Obama to Jesse Jackson in anticipation of the South Carolina results (January 26) and a subsequent complaint that Obama had “played the race card” (April 21). Whether or not the Clintons intended to portray Obama as “the black candidate,” as some charged, Bill Clinton’s comments were chewed over endlessly in the media, around watercoolers, and across dinner tables. Other supporters of Hillary Clinton also made controversial statements that surely cued group identity. Geraldine Ferraro said, “If Obama was a white man, he would not be in this position” (the remarks were first published on March 7). Just one day after the Ferraro story received national attention, an incendiary speech by Obama’s former pastor, Rev. Jeremiah Wright, was broadcast on the television show Good Morning America, leading to several days of fascination with Wright, his views, and his relationship with his most famous congregant. Obama responded on March 18 with a significant speech on race in America. A month later, Wright addressed

132

Chapter 6

the National Press Club, defending his earlier comments and offering some new firebombs (April 28); Obama’s response (distancing, then denunciation) unfolded over two days. At about the same time, the Pennsylvania primary results (April 22) directed much ink, bandwidth, and air time to the question of how Obama would fare among white voters against John McCain (see, e.g., Cohen 2008; Fram 2008; Nagourney 2008; Rede 2008; Thomma 2008; Weisman and Mosk 2008; Williams 2008). Thus a campaign that itself strove to be racially neutral acquired the “burden” of racial baggage. But that burden ultimately proved an asset, as the cues of group identity appear to have pulled black support from Clinton to Obama without pushing white support from Obama to Clinton. Black women who were conflicted between gender identity and racial identity increasingly resolved that conflict in favor of racial identity as the campaign progressed. If their critics were correct in charging the Clintons with cynical deployment of race as an issue, that tactic backfired. A more productive approach would likely have been to emphasize gender differences instead of racial differences. Indeed, during the course of the campaign, there were cues about gender identity in the form of commentary about Hillary Clinton’s temperament, appearance, and attire and the hostility that she purportedly evoked among many male voters and in the heated and angry response of many female activists and voters to this commentary. Still, gender cues ultimately were less prevalent than the recurrent, high-profile cues about racial identity. (In the end, most analysts conclude that Obama lost few votes simply because of his race).

THE ROLE OF RACE OVER TIME: JESSE JACKSON AND BARACK OBAMA This chapter has explored the ways the nominating process in 2008 shaped the influence of group identities on campaign narratives and voting behavior. The focus on the role of race in shaping the competition between Obama and Clinton raises the important question of how patterns of racial voting in 2008 compared to 1988, the last nominations contest with a viable black candidate. In 1988, Rev. Jesse Jackson won 10 nominating events, including six southern primaries (Alabama, the

The Rules and the Role of Race and Gender in 2008

133

District of Columbia, Georgia, Louisiana, Mississippi, and Virginia) and four caucuses dotted across the country (Alaska, Delaware, Michigan, and South Carolina). Of course, the future 44th president ultimately fared much better than the fiery civil rights leader. But the degree to which the constituencies of two black candidates who deliberately chose different campaign styles and messages overlapped can be explored. In particular, it is valuable to uncover the influence of Obama’s self-conscious effort to downplay race-conscious appeals on the breadth of his appeal. Two similarities between the two candidates’ state-by-state results are striking. First, consider support from white voters. Figure 6.5 depicts support for Jackson in 1988 and Obama in 2008 by state, and it is quite clear that Obama tended to do well among whites in the states where Jackson had done well among whites.14 The correlation between the state-level share of the vote for Jackson in 1988 and for Obama in 2008 is statistically significant and positive, but still relatively weak. There is no state where Jackson ran well and Obama floundered, and Obama did significantly better than Jackson in most states.

Figure 6.5. Support among white voters for Jesse Jackson (1998) and Barack Obama (2008).

134

Chapter 6

The similarity in the locales of strong support for Obama and Jackson presumably reflects a similar demographic pattern in white support for both men. Well-educated, liberal whites heavily favored both Obama in 2008 and Jackson in 1988, and the distribution of liberals across states probably did not change remarkably in the 20-year span. Illinois is an outlier, with Obama winning 56 percent of the white vote there versus the surprisingly small 8 percent that Jackson managed. Removing Illinois from the analysis yields a statistical result showing that Jackson’s white vote share predicts half of the variance in the white vote for Obama. The greater willingness of educated white liberals to support the two black candidates also underlies a second similarity in the pattern of support for their candidacies, namely their successes in the caucus format for delegate selection. In 2008, Obama won 11 out of 13 state Democratic caucuses (he also won two of four caucuses held in U.S. territories). In 1988, Jackson won four out of 13 caucuses, fewer than Michael Dukakis but still an impressive showing given that three of them were held north of the Mason-Dixon line. By contrast, Jackson did not win a single primary outside the South. As noted above, caucusgoers skew toward higher education levels compared to primary voters, so the overrepresentation of this group, highly supportive of black candidates running 20 years apart, almost certainly contributed to their common success in the caucus format. There is, however, a notable difference between Obama’s and Jackson’s state-by-state results. As discussed earlier (and depicted in figure 6.4b), support for Obama tended to increase among black voters (more so among women than men) over time. This was not the case during Jackson’s 1988 candidacy; indeed, over time his share of the black vote declined slightly. This trend is illustrated in figure 6.6, which plots Jackson’s share of the black vote (as estimated by ABC News exit polls) against the date of the Democratic primary. The trend line suggests that over the four-month duration of primary season, average black support for Jackson decreased by about 10 percentage points.15 This trend is difficult to explain in light of theories that point to the power of racial cues to boost feelings of ethnic identity and group pride. It therefore is unlikely that Jackson’s diminishing appeal to black voters was driven by changes in the visibility and intensity of race-

The Rules and the Role of Race and Gender in 2008

Figure 6.6.

135

Jesse Jackson’s share of the black vote over the primary season.

related cues, which seemed to foster Obama’s increased support among blacks during the 2008 campaign. Jackson’s campaign throughout included numerous prominent signals about the significance of race, such as the candidate’s own well-known personal story as a civil rights leader and his impassioned rhetoric about the need to attack racial inequality in America. It seems simply implausible that many black voters felt their group identity was becoming less salient over the course of this campaign. Increased group voting is also thought to be related to strategic calculations by voters. The standard expectation is that group members will be more “loyal” when candidates from their own group are perceived as viable, in the sense of having a chance to run strongly if not prevail. On the surface, Jackson’s “viability” increased rather than declined over time. As he accumulated primary and caucus victories, Jackson emerged as the main challenger to front-runner Michael Dukakis, seemingly making it illogical for blacks to have abandoned him because of concerns over wasting their votes.16 So the slight taperingoff in black support for Jackson in the mass public remains mysterious. It is true, though, that among political elites of both races, there never

136

Chapter 6

was a serious belief that he might win the Democratic nomination. Despite his strong electoral showing, he remained viewed as an insurgent candidate, someone with a strong ideological and symbolic appeal to the left of the party, a kind of McGovern redux whose opponent did not self-destruct. Thus, the Obama and Jackson campaigns differed in many ways, notwithstanding the shared race of the candidates. Obama’s campaign was well funded and based on a broad appeal to both Democrats and independents. Jackson was a candidate of a faction, both racial and ideological. The competitive environments they faced also were quite different. In 2008, Obama’s main opponent entered the year as a very strong favorite to win and had strong credentials among the black community. In 1988, none of the other candidates could rival Jackson’s appeal to blacks, and none entered as a presumptive front-runner. Consequently, Obama and Jackson faced different strategic imperatives, and the procedures in place for their respective efforts helped determine their ability to meet those imperatives. Black voters who started out well disposed toward the Clintons needed assurance that Obama had a real chance of winning the nomination; in general, blacks were not prepared to automatically offer him support, as many did for Jackson. At the same time, the Iowa caucus was an important signal of Obama’s appeal to whites and, much like John Kennedy’s win in Protestant West Virginia in 1960, an indication that his self-presentation as someone who looked beyond race could be effective in the general election. If Obama had faltered in Iowa, the compressed schedule may well have prevented him from recovering. If the first contest had been a primary—even the New Hampshire primary—instead of a caucus in a state neighboring Illinois, Obama might not have so quickly dispelled doubt about his viability. Jackson did not face such an early make-or-break moment, as the absence of a single strong white challenger for the hearts and minds of black voters meant that he did not have the same pressure to demonstrate broad viability. The field winnowed more slowly in 1988 than 2008, and as Democrats waited to see who the main white contenders would be, Jackson plugged along, picking up southern primaries with overwhelming black support and caucuses with help from educated white liberals. He thus earned considerable attention for his agenda, during both the nominations campaign and at the national convention.

The Rules and the Role of Race and Gender in 2008

137

By contrast, Obama’s strength was his ability to broaden the coalition of educated white liberals and black voters that had backed Jackson, particularly in primary states.

CONCLUSION The diversity of presidential candidates in 2008 was striking, and the ascendance of the Democrats Barack Obama and Hillary Clinton quite naturally focused attention on the historic aspect of their candidacies: after January, it was clear that the Democratic Party’s presidential nomination would be a “first” one way or the other. And while the party’s own procedures incorporate mechanisms, including some instituted for the 2008 cycle, to assure a voice for women and minorities in the selection, they did not materially help or hinder the black candidate or the female candidate in 2008 except perhaps indirectly (and even then minimally), because of African American voters’ overwhelming preference for Barack Obama. Obama surely enjoyed such strong support in part because of his race, but in a different field, any candidate so heavily favored by black voters would have benefited from the schedule changes, especially the promotion of South Carolina, that augmented black influence over the nomination. Sorting out the impact of the process on the outcome is challenging, even frustrating, because an individual presidential campaign cycle is such a singular event, or perhaps a singular collection of events. The DNC moved two delegate-selection events into the prewindow period. Two states acted on their own to join the early birds and were sanctioned for doing so. Many other states initiated schedule changes of their own, such that most Americans had already cast a ballot (or missed their opportunity to do so) by February 6. Some states switched the form of their nominating events. In the meantime, real campaigns were underway, and the happenings on one day affected those on the next. To disentangle the significance of any one aspect of the process from all the others is simply impossible. Still, there are two fairly clear lessons to be drawn. First, the net effect of each procedural change in 2008 was to enhance the influence of blacks, primarily through South Carolina’s prewindow primary, and Latinos, primarily through the pile-up of

138

Chapter 6

large-state events on Super Tuesday. For both groups, the impact of procedural changes was modest, but given the unusually close race between Clinton and Obama, they were potentially determinative.

NOTES 1. Republicans also held prewindow caucuses in Wyoming (January 5) and Maine (February 1). 2. To avoid cumbersome phrasing, I use “voters” to mean “voters and caucus attendees,” except when comparing the two groups, in which case the context makes the meaning of “voters” obvious. 3. In the minority empowerment literature, the work most relevant to the DNC-initiated schedule change considers how the creation of majorityminority districts affects minority participation (Brace, Handley, Niemi, and Stanley 1995). As with the Democratic Party’s calendar change, the creation of minority-dominated districts marks an opportunity for particular groups to carry more weight over a political decision, whether it is the election of an individual member of Congress or the tally in the much-watched, bellwether early contests in a presidential nomination. In their study of majority-minority congressional districts, Brace and colleagues did not find compelling evidence of an empowerment effect on turnout. 4. A superior measure of group turnout would be the most direct one, the number of individuals voting in each primary divided by the number eligible to do so. For any group, the numerator can be estimated from exit polls and statewide turnout data—in fact, approximately 291,500 blacks voted in the 2008 South Carolina primary but about 141,000 in 2004. The denominator, however, is not available, so one cannot determine how much of the net increase in 2008 is due to mobilization and how much is due to other factors, such as population shifts (was the black population of South Carolina significantly larger in 2008 than in 2004?) or changes in eligibility (did some significant number of blacks switch parties?). Technically, the measure of black turnout used herein should actually be called “the proportion of the primary electorate that is black”; however, in the interest of conciseness, I will refer to it as “black turnout.” It is possible that the black share of the electorate could increase by 8 points even if the turnout rate among blacks and whites were the same if there were significant population or partisan shifts in South Carolina between 2004 and 2008, but this seems highly unlikely.

The Rules and the Role of Race and Gender in 2008

139

5. While at least one study in the minority empowerment field has documented increased turnout among blacks when black candidates are on the ballot (Vanderleeuw and Utter 1993), one cannot be certain that this finding would generalize to Latinos, a presidential primary, or a presidential caucus. 6. If the potential for an enhanced voice motivated blacks to vote in South Carolina’s primary, it could also have mobilized blacks in late-voting states. After a long primary season, the Democratic nomination was still unresolved in May, giving states like North Carolina and Indiana the potential to tip the scale once and for all. This would, of course, have resulted not from the official procedures but from the way the campaign evolved. Unfortunately, the lack of 2004 exit polls from North Carolina and Indiana (the nomination had been determined by the time those state primaries were held) precludes a comparison of group turnout in those states. 7. Of course, data on the daily fundraising gap is available for the entire period of the nomination contest, but to keep focus on the impact of South Carolina and Nevada, only the first two months of 2008 are depicted. 8. Black population statistics are from the U.S. Census Bureau, 2007 American Community Survey. 9. The prolonged battle for the Democratic nomination essentially gave late-voting states more influence in 2008 than in 2004. Thus, holding the same position in the calendar did not necessarily mean holding the same amount of influence, as discussed earlier. Nonetheless, there is no state other than Nevada with a large Latino population that stands out as having significantly more or less influence in 2008 compared to 2004. 10. Obama’s total excludes funds earmarked for the general election (see Campaign Finance Institute 2008b). 11. Because federal candidates are not obligated to report any information about supporters who contribute less than $200, researchers’ ability to survey the small-donor population is limited by individual candidates’ willingness to voluntarily provide lists of small donors. The survey reported by Graf and colleagues (2006) did not include any information about contributors to either major-party nominee (George W. Bush and John Kerry) or to Democrat Howard Dean, who raised considerable funds from small donors. 12. The gender makeup of Obama’s bundlers was reported in Schouten (2008). Hillary Clinton’s bundlers estimated by the author from data provided by Public Citizen at www.whitehouseforsale.org. 13. Later, other observers made similar claims about Bill Clinton, one of which drew significantly more attention than Steele’s essay in Time magazine for its crudeness.

140

Chapter 6

14. This tendency is slightly obscured by marked differences in the competitive environments each candidate faced. Beginning on Super Tuesday, Obama’s only competitor for the white vote in all of these states was Hillary Clinton, whereas Jackson faced as few as two opponents (in New Jersey, where Senator Al Gore was the only other remaining challenger to the eventual nominee, Governor Michael Dukakis) and as many as six (in the early contests). For this reason, figure 6.5 only depicts states in which Jackson faced the full slate of six opponents: Dukakis, Gore, Senator Paul Simon, Governor Bruce Babbitt, Representative Dick Gephardt, and Senator Gary Hart. 15. The trend line was fitted using linear regression, with weights for number of black survey respondents in each state. Three of the obvious outliers included small states with relatively few black votes and, therefore, fewer than 20 black respondents in the exit pools, resulting in large margins of error. 16. Indeed, there is no reason to think they did so; the beneficiary of Jackson’s slight dip in black support was not an alternative challenger but Dukakis himself, who certainly did not need help from black voters to fend off Al Gore, the last white challenger in the race. Dukakis’s share of the black vote increased over time by nearly the amount that Jackson’s share decreased.

7 THE WAY WE CHOOSE PRESIDENTIAL NOMINEES: PROBLEMS AND PROSPECTS David E. Price

he 2008 presidential nomination was full of surprises, even for party rules-writing veterans fully aware of the law of unintended consequences. In the fall of 2007, I spoke at a Rutgers University forum, highlighting three things that I thought were wrong, or were likely to go wrong, with the system.

T

• It started too soon—but, ironically, might well end too abruptly. • The contests were too concentrated—front-loaded rather than paced through February, March, April, and May. • The early concentration began to look like a national primary, with an accentuation of the impact of big money and big media. Well, how about that? The Democratic contests began on January 3 (too soon!) but extended through the entire nomination season, with the outcome in some doubt until early May. Most people probably would not say that even the Republican contest ended “too abruptly.” The contests were indeed concentrated—there were long stretches with no primaries or caucuses—but the few states that waited played a decisive role. I confess to great satisfaction at this: the states, including North Carolina, that exercised restraint and did not rush to the front of the season were at the heart of the Democratic campaign. May 6, the date of the North Carolina and Indiana primaries, was the Tuesday that sealed Senator Obama’s victory. But however you might want to characterize this unusual season, the term “national primary” does not come to mind! 141

142

Chapter 7

So I’ll eat my share of crow. But I also want to recall what the Commission on Presidential Nomination Timing and Scheduling, which I co-chaired in 2005 with former Secretary of Labor Alexis Herman, tried to do. Our best-laid plans were often frustrated, but the concerns on which our actions were based still offer positive guidance going forward. My comments will focus mainly on the Democratic Party, but many will be more broadly applicable. Let me first state what is perhaps obvious: party reforms are not creating a system from the ground up but are modifying an inherited system of decentralized local and state party politics. Our patchwork system of state-by-state contests has evolved from a system of successive precinct, county, and state conventions that then culminated in national conventions. There, the delegates elected through these earlier processes gathered to make a decision that often wasn’t predetermined at all and in fact often went on for several days through multiple ballots. Nowadays, caucus and convention decision making has largely been replaced by primary elections, and voting at the convention contains few surprises—it is tightly bound to the outcome of the primaries. A series of party reforms have thoroughly transformed the old convention system. But what remains is a state-by-state sequence of contests. In some ways, they represent a substitute or surrogate for the iterative process that went on in successive conventions, with an opportunity for candidacies to rise and fall, for a certain testing of candidates to go on, for the electorate in various states to assume complementary roles in winnowing the field. That, at any rate, is how it is supposed to work—quite different from the one-shot, big-bang event a national primary would represent. Why not have a national primary? Are we going through all these twists and turns to preserve the vestiges of an outmoded system? A national primary, after all, would be quick and simple. Every state would vote on the same day, and every voter would have an equal say. Most Americans say they favor such a system. But despite these virtues, we should not be seduced by the idea of a national primary. It would result in big money and big media having even more influence than they do now. It would deal a devastating blow to grassroots and state-level politics. And it would result in a less thorough consideration of candidates, and certainly a less correctable, cumulative judgment, than does our present system, whatever its flaws.

The Way We Choose Presidential Nominees

143

In stressing the historic origins of our system of sequential contests, I do not want to minimize how far we have come toward what might be termed a plebiscitary system—one based on the popular vote, where other ways of shaping the nomination outcome are greatly marginalized and even delegitimized. Perhaps I can use the current “superdelegate” discussion to illustrate this. This will also give me a chance to look back briefly at the reform commission with which I was most closely involved (as staff director), the Hunt Commission of 1981–1982. The Hunt Commission, chaired by North Carolina’s Governor Jim Hunt, was the fourth of the four reform commissions that followed the chaotic 1968 Democratic National Convention. In most respects, it finetuned the work of its predecessor commissions: McGovern-Frazer, Mikulski, and Winograd. But in one respect, it set out to reverse a major unintended consequence of reform: the elimination of top elected officials from the convention. The reformers, in their zeal to break the perceived power of party bosses and produce a convention reliably tied to the popular vote, produced a convention of candidate activists—delegates slated by candidates who were allocated according to primary or caucus outcomes. This largely eliminated elected officials and even many top party leaders for two reasons: they were unwilling to run against their own constituents to become delegates, and they were often either unwilling to make an early pledge to a presidential candidate or unable to get on a candidate’s slate. It also, we perceived, contributed to a certain detachment on the part of members of Congress and other elected officials from the Democratic ticket and the fall campaign. Therefore, the Hunt Commission added slots for party leader/elected official delegates—PL/EOs—to each state’s delegation and determined that most of them could remain unpledged: hence the misleading term “superdelegate.” How did we envision their decision-making role? Not, surely, as king makers or queen makers, much less as some sort of rump convention. There was, however, some discussion about the relatively rare circumstances—a deadlocked contest, closely divided platform questions, a situation when a candidate became disabled or disgraced— when the decision-making capacity the PL/EOs would bring to the convention might prove valuable. Now, to be sure, this discussion took place in a context of concern, among some but by no means all Hunt Commission members, that the

144

Chapter 7

convention had become too bound to candidates and had lost all deliberative capacity. But the Hunt Commission took only marginal steps to unbind convention delegates, and we certainly did not look to the infusion of PL/EOs to do more than slow this trend. The 26 ensuing years marked a further fading of any memory of or yearning for a “deliberative convention.” Today, any outcome that had PL/EO delegates reversing even a close primary season outcome, as reflected in the distribution of pledged delegates, would likely be widely seen as illegitimate. That is what I mean by a “plebiscitary” convention. So, I am often asked, what should a so-called superdelegate consider in casting his or her vote? My answer is that a number of considerations are legitimate: one’s own judgment based on knowledge of and experience with the candidates; the preference of voters in one’s district or state; the potential of the candidates to win in the general election. But in the end, the PL/EO delegates will be more likely to resolve than to create uncertainty, to consolidate behind the winner among the pledged delegates and to ratify that choice. The party will normally be better off if this verdict is clear several months before the national convention convenes. Our 2005 commission had a narrower charge than the Hunt Commission, as its official name indicates: Presidential Nomination Timing and Scheduling. The commission was appointed by the DNC chair to head off a threat by leaders from Michigan to upend the 2004 calendar because of their unhappiness at Iowa and New Hampshire’s monopoly of the early season, before other states were allowed to schedule their contests. So it was foreordained that the first question we would deal with was the preseason’s bellwether contests, which we chose to retain but to diversify demographically and regionally with the addition of a few more states—eventually determined to be South Carolina and Nevada. As a second step, we provided bonus delegates, to induce states to refrain from moving to the start of the regular season on or shortly after February 5. This largely failed, just as a milder version had failed earlier for the Republicans, although some 10 states were awarded bonus delegates in 2008 for moving their contests later or leaving them in place in the latter part of the season.

The Way We Choose Presidential Nominees

145

Our third major recommendation urged Congress to renew and modernize the virtually defunct presidential public-financing system. At present, the system’s near collapse has interacted with the compressed calendar to give private fundraising (and the fundraisers) a more prominent role than ever. And finally, we recommended interparty negotiations to set a later starting date for the entire 2012 process. Republican practice was to set their rules for a given election at the prior national convention. Democrats would therefore need to seek bipartisan agreement, for they would effectively be bound by whatever date the Republicans locked in place at their 2008 convention. Our plan to scatter four diverse small-state contests between midJanuary and February 5 was upended when Florida and Michigan leapt into the early season. The approved early states then leapfrogged the leapfroggers, pushing the Iowa caucuses to January 3. The DNC refused to recognize the Florida and Michigan contests, and the Democratic candidates signed an agreement—urged upon them by the approved early states—not to campaign there. The party’s decision was fully justified, yet in politics there is often a danger of cutting off one’s nose to spite one’s face. So the DNC and the Michigan and Florida state parties held extensive discussions about how to seat the delegations and bring these critical states back into the fold—discussions both complicated and given greater urgency by the closeness of the Obama-Clinton contest. Allocating delegates according to the January primary votes was out of the question, and both states ruled out revotes. By the time the Rules Committee reached a compromise on the allocation of delegates to the candidates, these votes were no longer critical to the outcome. Michigan and Florida delegates were initially given only a half vote each, a penalty comparable to that imposed on these states by the Republicans, but their full voting rights were restored as the convention began. Longer-term questions remain. How can the parties prevent the preseason from again being swamped? Or should the preseason simply be eliminated, which is clearly the goal of at least one of the leapfroggers, Michigan? The early season chaos was accompanied by extraordinary regularseason front-loading. Twenty-two states rushed to February 5; the votes

146

Chapter 7

of more than 50 percent of the pledged delegates to both parties’ conventions were thus determined as of the official season’s first day. This was hardly the nomination schedule that our commission or the leaders of either party wanted. We sought changes that would involve a more diverse group of citizens in vetting our candidates, preserve the grassroots components of the process, pace the contests more evenly, simulate a process of deliberation and cumulative judgment, and help choose a strong and credible nominee. The verdict on these process goals is mixed at best, and that, plus other accumulated dissatisfactions from the 2008 experience, has made another round of reform quite likely. There is no question that the free enterprise of 100+ state parties and 50 legislatures have failed to produce an orderly or optimal outcome—we have our own “tragedy of the commons,” a case of political market failure. I’ve become more receptive than I used to be to some sort of federally legislated solution—not a national primary, but perhaps rotating clusters of primaries—but I am acutely aware of the pitfalls: not only the extraordinary difficulty of achieving congressional consensus on such a matter, but the very real possibility of judicial rejection of such intrusion. So it is still preferable that the parties fix themselves. Maybe this year’s combination of sanctions imposed on the leapfrog states and virtue rewarded for the late-voting states will make for altered calculations of advantage by state parties and legislatures next time. But the parties will still need to evaluate the 2008 results and reassess the limited incentives and sanctions they control and how they might be deployed. That is not to say another full-blown commission is required, but the Democrats have determined to have one anyway. Senator Clinton’s forces came to the 2008 convention determined to do something about party caucuses, which had delivered repeated wins for Senator Obama, while the Obama forces wanted to modify the superdelegate system. Rather than adopt these and other rules changes directly, the convention’s Rules Committee proposed a 37-member Democratic Change Commission to recommend changes in the delegate selections rules for 2012. The commission is to develop “new specific criteria” for organizing and standardizing party caucuses and broadening caucus participation. Some no doubt will wish to go further, discouraging caucuses in favor of primaries. As for the unpledged PL/EO delegates, the conven-

The Way We Choose Presidential Nominees

147

tion resolution called for a “significant reduction” in their number, with no recognition of the rationale for creating those slots in the first place. The Democratic resolution also reflected our commission’s recommendation of a later start date, designating March 6 as the first day of the regular season in 2012. This was discussed with the Republican National Committee leadership, who for a time held out hope for far more ambitious calendar revisions. Some of them were still smarting from Karl Rove and George W. Bush’s rebuff at their 2000 convention, when they were about to adopt the “Delaware plan” (a small-states-first plan that had been seriously discussed in both parties). In early 2008, the RNC proposed what they called a “modified Ohio plan,” which provided for preseason contests in New Hampshire, Iowa, Nevada, and South Carolina; let a cluster of other small states votes next; and then sequenced three further clusters of states, rotating their places in the sequence, election to election. In the end, the 2008 Republican National Convention went in another, largely unanticipated direction, authorizing a 15-member reform commission of its own. Republicans also backed off the later opening date agreed upon with Democrats, reverting to early February. But in abandoning the GOP’s practice of locking in rules at the prior national convention, the party opened up the possibility of using their commission to alter the starting date and much else besides. The creation of reform commissions in both parties guarantees renewed assessment of the nomination process and opens up new possibilities for interparty collaboration. It also will pose formidable challenges— not only to achieve agreement between the parties but also to secure the cooperation of the state parties and legislatures upon which a well-ordered calendar largely depends. Political stakeholders and analysts in the media and academia will be well advised to stay tuned. These are arcane matters, but highly consequential for democratic participation and legitimacy, leadership, and governance.

NOTE This chapter is a revised version of the keynote address Representative Price gave at the Berkeley Conference on “Nominating the President: Its Evolution and Effects” in April 2008.

REFERENCES

Adkins, Randall E., and Andrew J. Dowdle. 2004. “Bumps in the Road to the White House: How Influential Were Campaign Resources to Nominating George W. Bush?” Journal of Political Marketing 3(4): 1–27. ———. 2001a. “Is the Exhibition Season Becoming More Important to Forecasting Presidential Nominations?” American Politics Research 29(3): 283–88. ———. 2001b. “How Important Are Iowa and New Hampshire to Winning Post-Reform Presidential Nominations?” Political Research Quarterly 54(2): 431–44. ———. 2000. “Break Out the Mint Juleps in New Hampshire: A Forecasting Model of the Presidential Primaries.” American Politics Quarterly 28(2): 251–69. Alexander, Lamar. 1996. “Off with the Limits.” Campaigns and Elections, October–November, 32–35. Ambinder, Marc. 2008. “Ohio Plan Advances in RNC Rules Deliberation.” Atlantic, August 19, http://marcambinder.theatlantic.com/archives/. Arnoldy, Ben. 2008. “Latino Vote Critical for Clinton on Super Tuesday.” Christian Science Monitor, February 4. Atkeson, Lonna Rae. 1999. “‘Sure, I Voted for the Winner!’: Overreport of the Primary Vote for the Party Nominee in the National Election Studies.” Political Behavior 21(3): 197–215. ———. 1998. “Divisive Primaries and General Election Outcomes: Another Look at Presidential Campaigns.” American Journal of Political Science 42 (January): 256–71. Ayres, B. Drummond, Jr. 1999. “Presidency in 2000: $2,893 an Hour.” New York Times, January 12, A17. Balz, Dan. 2008. “Obama Looking to Diminish Superdelegates.” Washington Post, August 20, http://voices.washingtonpost.com. 149

150

References

Barone, Michael. 2008a. “South Carolina Reshapes the Race.” U.S. News and World Report, www.usnews.com. ———. 2008b. “A Super Fight Down the Road,” www.aei.org. Bartels, Larry M. 1988. Presidential Primaries and the Dynamics of Public Choice. Princeton, N.J.: Princeton University Press. Bernstein, Robert A. 1977. “Divisive Primaries Do Hurt: U.S. Senate Races 1956–1972.” American Political Science Review 71: 540–45. Bobo, L., and F. D. Gilliam. 1990. “Race, Sociopolitical Participation, and Black Empowerment.” American Political Science Review 84(2): 377–93. Born, Richard. 1981. “The Influence of House Primary Election Divisiveness on General Election Margins, 1962–1976.” Journal of Politics 43: 640–61. Brace, K., L. Handley, R. G. Niemi, and H. W. Stanley. 1995. “Minority Turnout and the Creation of Majority-Minority Districts.” American Politics Quarterly 23(2): 190–203. Brainerd, Kristen. 1999. “With an Eye toward Speakership, Gephardt May Soon Spurn Presidential Bid.” Congressional Quarterly Weekly Report, January 23, 197–98. Brazile, Donna. 2007. “Why We Stood Up to Florida.” Washington Post, September 27, A27. Brewer, Marilynn B. 2001. “The Many Faces of Social Identity: Implications for Political Psychology.” Political Psychology 22(1): 115–25. Busch, Andrew E., and James W. Ceaser. 1996. “Does Party Reform Have a Future?” In In Pursuit of the White House: How We Nominate Our Presidents, ed. William G. Mayer. Chatham, N.J.: Chatham House. Campaign Finance Institute. 2007. “Big $1,000+ Donations Supply 79% of Candidates’ Early Money.” April 16, www.cfinst.org. ———. 2008a. “After Holding Financial Advantage in Primaries, Obama Likely to Achieve Only Parity with McCain in General Election.” September 26, www.cfinst.org. ———. 2008b. “Obama’s Small Contributions Surged in June, but McCain’s Party-Based Strategy Gave the GOP Side a Combined Cash Advantage on June 30.” www.cfinst.org. ———. 2008c. “Reality Check: Obama Received about the Same Percentage from Small Donors in 2008 as Bush in 2004; Obama Also Raised 80% More from Large Donors than Small, Outstripping All Rivals and Predecessors.” November 24, www.cfinst.org. Carleton, William G. 1957. “The Revolution in the Presidential Nominating Convention.” Political Science Quarterly 72(2): 224–40.

References

151

Center for Responsive Politics. 2008. “Bundlers.” www.opensecrets.org. Cohen, Marty, David Karol, Hans Noel, and John Zaller. 2008. The Party Decides: Presidential Nominations before and after Reform. Chicago: The University of Chicago Press. Cohen, Richard. 2008. “The Race Issue, Still.” Washington Post, April 8, A19. Cohen, Sarah. 2008. “Small Donations Create Big Obama Donors.” The Trail Blog, Washington Post, October 23, http://voices.washingtonpost.com. Corrado, Anthony. 2006. “Party Finance in the Wake of BCRA: An Overview.” In The Election after Reform: Money, Politics, and the Bipartisan Campaign Reform Act, ed. Michael Malbin. Lanham, Md.: Rowman & Littlefield. Daily Kos. 2008. “Public Financing in the Primary Is Dead.” August 21, www.dailykos.com. Democratic National Convention. 2004. “Resolution Establishing a Commission on Presidential Nomination Timing and Scheduling.” DNC Office of Party Affairs and Delegate Selection. 2005. Report of the Commission on Presidential Nomination Timing and Scheduling. Washington, D.C.: Democratic National Committee. Dolan, Kathleen A. 2004. Voting for Women: How the Public Evaluates Women Candidates, Dilemmas in American Politics. Boulder, Colo.: Westview. Dunn, Charles W. 2008. “Cherry Picking and Leveraging in a Civil War.” Chuck Dunn Report, January 22, www.thechuckdunnreport.blogspot.com. Elving, Ronald D. 1995. “Gramm Launches Campaign Stressing Familiar Themes.” Congressional Quarterly Weekly Report, February 25, 630. Federal Election Commission. 2008. “Public Funding of Presidential Elections.” Published in August 1996, updated February 2008, http://www.fec.gov. Fenno, Richard F., Jr. 1978. Home Style: House Members in Their Districts. New York: Harper-Collins. Fleishman, Joel L., and Pope McCorkle. 1984. “Level-Up Rather than LevelDown: Toward a New Theory of Campaign Finance Reform.” Journal of Law and Politics 1: 211. “For 2012, the Same Old Mess.” 2008. USA Today, September 3, 14A. Forsyth, Jim. 2008. “Huckabee: A Deadlocked Convention Is My Goal.” http://radio.woai.com. Forum. 2008. April 9, www.bepress.com/forum. FOXNews.com. 2008. “Florida Exit Polls: Key Demographics Push McCain to Victory 2008.” http://elections.foxnews.com. Fram, Alan. 2008. “Polls: Race Helps Clinton with Whites.” Associated Press Online, April 7, www.ap.org.

152

References

Franceschet, Susan, and Jennifer M. Piscopo. 2008. “Gender Quotas and Women’s Substantive Representation: Lessons from Argentina.” Politics and Gender 4: 393–425. Geer, John G. 1989. Nominating Presidents. New York: Greenwood. “GOP Adopts Rules, Platform on Both Sides of McCain.” 2008. AP News, September 1, http://townhall.com. Graf, Josef, Grant Reeher, Michael J. Malbin, and Costas Panagopoulos. 2006. Small Donors and Online Giving: A Study of Donors to the 2004 Presidential Campaigns. Washington, D.C.: Graduate School of Political Management, George Washington University. Green, John C. 2006. “Financing the 2004 Presidential Nomination Campaigns.” In Financing the 2004 Elections, ed. David B. Magleby, Antony Corrado, and Kelly D. Patterson. Washington, D.C.: Brookings Institution. Green, John C., and Nathan S. Bigelow. 2006. “The 2000 Presidential Nominations: The Costs of Innovation.” In The Election after Reform: Money, Politics, and the Bipartisan Campaign Reform Act, ed. Michael Malbin. Lanham, Md.: Rowman & Littlefield. Hacker, Andrew. 1965. “Does a Divisive Primary Harm a Candidate’s Election Chances?” American Political Science Review 59: 105–10. Halloran, Liz. 2008. “Too Early, Too Long, Too Divisive.” U.S. News and World Report, July 7–July 14, 21. Hallow, Ralph Z. 2008. “GOP Panel Rejects Proposal to Spread Out 2012 Primaries.” Washington Times, August 28, A12. Hanson, Gordon S. 2000. “Out of Cash: U.S. Presidential Pre-Nominations.” Party Politics 6(1): 47–59. Hasen, Richard L. 2003. The Supreme Court and Election Law: Judging Equality from Baker v. Carr to Bush v. Gore. New York: New York University Press. ———. 2008a. “Beyond Incoherence: The Roberts Court’s Deregulatory Turn in FEC v.Wisconsin Right to Life.” Minnesota Law Review 92: 1064. ———. 2008b. “Political Equality, the Internet, and Campaign Finance Regulation.” Forum 6(1), http://electionlawblog.org. Haynes, Audrey A., Paul-Henri Gurian, Michael H. Crespin, and Christopher Zorn. 2004. “The Calculus of Concession: Media Coverage and the Dynamics of Winnowing in Presidential Nominations.” American Politics Research 32(3): 310–37. Huddy, Leonie. 2003. “Group Identity and Political Cohesion.” In Oxford Handbook of Political Psychology, ed. David O. Sears, Leonie Huddy, and Robert L. Jervis. New York: Oxford University Press. Hutchings, Vincent L., and LaFleur Stephens. 2008. “African American Voters and the Presidential Nomination Process.” In The Making of the Presidential

References

153

Candidates 2008, ed. William G. Mayer. Lanham, Md.: Rowman & Littlefield. Institute for Politics, Democracy, and the Internet. 2006. “Small Donors and Online Giving: A Study of Donors to the 2004 Presidential Campaigns,” www.ipdi.org. Issenberg, Sasha. 2008. “Candidates Spread Thin in the Push to Feb. 5.” Boston Globe, January 9, A1. Joe, Wesley, Michael Malbin, Clyde Wilcox, Peter Brusoe, and Jamie Pimlott. 2008. “Do Small Donors Improve Representation? Some Answers from Recent Gubernatorial and State Legislative Elections.” In Annual Meeting of the American Political Science Association. Boston, Mass. Karol, David. 2009. Party Position Change in American Politics: Coalition Management. New York: Cambridge University Press. Kenney, Patrick J. 1988. “Sorting Out the Effects of Primary Divisiveness in Congressional and Senatorial Elections.” Western Political Quarterly 41: 759–77. Kenney, Patrick J., and Tom W. Rice. 1988. “Presidential Prenomination Preferences and Candidate Evaluations.” American Political Science Review 82: 1309–20. ———. 1987. “The Relationship between Divisive Primaries and General Election Outcomes.” American Journal of Political Science 31 (February): 31–44. ———. 1984. “The Effect of Primary Divisiveness in Gubernatorial and Senatorial Elections.” Journal of Politics 46: 904–15. Kirkpatrick, David D. 2007. “Edwards Embrace of Public Money May Be Limited.” New York Times, September 29, www.nytimes.com. Koch, Jeffrey W. 1994. “Group Identification in Political Context.” Political Psychology 15(4): 687–98. Kuhnhenn, Jim. 2008. “Obama Spends $55 Million in July, McCain $32M,” August 20, www.usatoday.com. Lengle, James I. 1980. “Divisive Presidential Primaries and Party Electoral Prospects, 1932–1976.” American Politics Quarterly 8: 261–77. Lengle, James I., Diana Owen, and Molly W. Sonner. 1995. “Divisive Nominating Mechanisms and Democratic Party Electoral Prospects.” Journal of Politics 57: 370–83. Lester, Will. 2006. “Bayh Has $9.5 Million in Campaign Funds.” Associated Press Online, January 28, www.ap.org. Loevy, Robert. 2000. “Toward a Sports Playoff System of Presidential Primaries and Caucuses.” In Nominating Future Presidents:A Review of the Republican Process. Washington, D.C.: Republican National Committee. Lowenstein, Daniel Hays, Richard L. Hasen, and Daniel P. Tokaji. 2008. Election Law: Cases and Materials. 4th ed. Durham, N.C.: Carolina Academic Press.

154

References

Magleby, David B., and Kelly D. Patterson. 2006. “Stepping Out of the Shadows? Ground-War Activity in 2004.” In The Election after Reform: Money, Politics, and the Bipartisan Campaign Reform Act, ed. Michael Malbin. Lanham, Md.: Rowman & Littlefield. Malbin, Michael J. 2006. “A Public Funding System in Jeopardy: Lessons from the Presidential Nomination Contest of 2004.” In The Election after Reform: Money, Politics, and the Bipartisan Campaign Reform Act, ed. Michael Malbin. Lanham, Md.: Rowman & Littlefield. Marshall, Thomas R. 1978. “Turnout and Representation: Caucuses versus Primaries.” American Journal of Political Science 22(1): 169–82. Mayer, William G. 2004. “The Basic Dynamics of the Contemporary Nomination Process: An Expanded View.” In The Making of the Presidential Candidates 2004, ed. William G. Mayer. Lanham, Md.: Rowman & Littlefield. ———. 2003. “Forecasting Presidential Nominations or, My Model Worked Just Fine, Thank You.” PS: Political Science and Politics 36(2): 153–57. ———. 2001. “The Presidential Nominations.” In The Election of 2000, ed. Gerald M. Pomper. New York: Chatham House. ———. 1996. “Forecasting Presidential Nominations.” In In Pursuit of the White House: How We Select Our Presidential Nominees, ed. William G. Mayer. Chatham, N.J.: Chatham House. Mayer, William G., and Andrew E. Busch. 2004. The Front-Loading Problem in Presidential Nominations. Washington, D.C.: Brookings Institution. McMullen, Jody L., and Barbara Norrander. 2000. “The Gender Gap in Presidential Nominations.” In In Pursuit of the White House, 2000, ed. William G. Mayer. Chatham, N.J.: Chatham House. Morain, Dan. 2008. “From Small Donors Come Big Rewards.” Los Angeles Times, February 14, www.latimes.com. Moran, Jack, and Mark Fenster. 1982. “Voter Turnout in Presidential Primaries: A Diachronic Analysis.” American Politics Quarterly 10: 453–76. Mosk, Matthew. 2008. “McCain Got Loan by Pledging to Seek Public Funds.” Washington Post, February 16, www.washingtonpost.com. Mosk, Matthew, and Sarah Cohen. 2008. “Big Donors Drive Obama’s Money Edge.” Washington Post, October 22, www.washingtonpost.com. Nagourney, Adam. 2008. “For Democrats, Questions over Race and Electability.” New York Times, April 24, 1. Nevada Democratic Party. 2007. “Final Turnout for Nevada Caucus: 117,599 Democrats.” www.nvdemscaucus.com. Newport, Frank. 2008. “Obama Gains among Former Clinton Supporters.” Gallup Poll, September 2, 2008, www.gallup.com.

References

155

Norrander, Barbara. 2006. “The Attrition Game: Initial Resources, Initial Contests, and the Exit of Candidates during the U.S. Presidential Primary Season.” British Journal of Political Science 36 (July): 487–507. ———. 2003. “The Intraparty Gender Gap: Differences between Male and Female Voters in the 1980–2000 Presidential Primaries.” PS: Political Science and Politics 36(2): 181–86. ———. 2000a. “The End Game in Post-Reform Presidential Nominations.” Journal of Politics 62 (November): 999–1013. ———. 2000b. “Is the Nomination Process Broken?” Nominating Future Presidents: A Review of the Republican Process. Washington, D.C.: Republican National Committee, 133–38. ———. 1986a. “Selective Participation: Presidential Primary Voters as a Subset of General Election Voters.” American Politics Quarterly 14: 35–53. ———. 1986b. “Correlates of Vote Choice in the 1980 Presidential Primaries.” Journal of Politics 48: 156–67. Norrander, Barbara, and Gregg W. Smith. 1985. “Type of Contest, Candidate Strategy, and Turnout in Presidential Primaries.” American Politics Quarterly 13: 28–50. Ohlemacher, Stephen. 2008. “Democrats to Review Nominating Process.” AP News, August 20, http://ap.google.com. Ornstein, Norman J., and Thomas E. Mann. 2008. “Delegates of Steel.” www.aei.org. Paolino, Phillip. 1995. “Group-Salient Issues and Group Representation: Support for Women Candidates in the 1992 Senate Elections.” American Journal of Political Science 39(2): 294–313. Parkin, Michael, and Rachel Welsh. 2008. “Minority Voters, Minority Candidates: Race and Gender in the 2008 Democratic Primary.” In Annual Meeting of the International Association of Political Psychology. Paris, France. Pew Research Center for the People and the Press. 2008. “Internet’s Broader Role in Campaign 2008: Social Networking and Online Videos Take Off.” January 11, http://people-press.org. Pierson, James E., and Terry B. Smith. 1975. “Primary Divisiveness and General Election Success: A Reexamination.” Journal of Politics 37: 555–62. Pitkin, Hanna Fenichel. 1967. The Concept of Representation. Berkeley: University of California Press. Political Intelligence. 2008. “Obama Reaches 2 Million Donors.” Boston Globe, August 14, www.boston.com. Political Pulse:The Newsletter of California Politics and Government. 2008. February 15.

156

References

Polsby, Nelson W. 1983. The Consequences of Party Reform. New York: Oxford University Press. Pooley, Eric. 1998. “The Priest at the Party.” Time, April 27, 36–38. Price, David E. 1984. Bringing Back the Parties. Washington, D.C.: CQ Press. Rapoport, Ronald B., and Walter J. Stone. 2000. “To: RNC Advisory Commission on the Presidential Nominating Process.” Nominating Future Presidents: A Review of the Republican Process. Washington, D.C.: Republican National Committee, 139–47. Rede, George. 2008. “Results from Pennsylvania Primary Reflect Race Chasm.” Sunday Oregonian, April 27, E05. Reiter, Howard. 1985. Selecting the President:The Nominating Process in Transition. Philadelphia: University of Pennsylvania Press. Roth, Bennett. 2008. “Latino Vote Is the Ticket for Super Tuesday.” Houston Chronicle, February 3. Rothenberg, Stuart. 2008. “Forget Expectations—Now Delegates Are What Really Matters.” April 25, www.realclearpolitics.com. Schatz, Amy. 2008. “Nielsen: About Three Million Got Obama Text.” Wall Street Journal, Washington Wire, August 25, http://blogs.wsj.com. Schouten, Fredreka. 2008. “In Fundraising Ranks, a Gender Gap Is Showing.” USA Today, July 30, www.usatoday.com. Shephard, Scott. 2008. “Hope and History Meet: His Campaign Rewriting the Rules; Long-Haul Strategy, Internet Savvy and Grass-Roots Surge Outmaneuvered Rival.” Atlanta Journal-Constitution, June 8, 3G. Sigelman, Lee, and Carol K. Sigelman. 1982. “Sexism, Racism, and Ageism in Voting Behavior: An Experimental Analysis.” Social Psychology Quarterly 45(4): 263–69. Simkins, Chris. 2008. Latino Voters Key to Super Tuesday. January 31, www.voanews.com. Southwell, Priscilla L. 1986. “The Politics of Disgruntlement: Nonvoting and Defection among Supporters of Nomination Losers, 1968–1984.” Political Behavior 8: 81–95. Steele, Shelby. 2007. “The Identity Card.” Time, November 30. Steger, Wayne P. 2000. “Do Primary Voters Draw from a Stacked Deck? Presidential Nominations in an Era of Candidate-Centered Campaigns.” Presidential Studies Quarterly 30(4): 727–55. Steger, Wayne P., Andrew J. Dowdle, and Randall E. Adkins. 2004. “The New Hampshire Effect in Presidential Nominations.” Political Research Quarterly 57(3): 375–90. Stone, Walter J. 1986. “The Carryover Effect in Presidential Elections.” American Political Science Review 80 (March): 271–79.

References

157

Stone, Walter J., Lonna Rae Atkeson, and Ronald B. Rapoport. 1992. “Turning On or Turning Off? Mobilization and Demobilization Effects of Participation in Presidential Nomination Campaigns.” American Journal of Political Science 36: 665–91. Tayler, Letta. 2008. “Latinos May Hold Key for Democrats on Super Tuesday.” Newsday, February 2. Thomma, Steven. 2008. “Obama Struggling to Win White Voters.” Contra Costa Times, April 23. Vanderleeuw, James M., and Glenn H. Utter. 1993. “Voter Roll-Off and the Electoral Context—A Test of Two Theses.” Social Science Quarterly 74(3): 664–73. Volokh, Eugene. 1991. “Cheap Speech and What It Will Do,” Yale Law Journal 104: 1805. Wayne, Stephen J. 2001. The Race to the White House 2000:The Politics of Presidential Elections. Post-election edition. New York: St. Martin’s. ———. 2000. “A Proposal to Reform the Presidential Nomination Process.” Nominating Future Presidents: A Review of the Republican Process. Washington, D.C.: Republican National Committee, 129–31. Weisman, Jonathan, and Matthew Mosk. 2008. “Party Fears Racial Divide: Attacks Could Do Lasting Harm, Democrats Say.” Washington Post, April 26, A01. Westlye, Marck C. 1991. Senate Elections and Campaign Intensity. Baltimore: Johns Hopkins University Press. Williams, Joseph. 2008. “Obama Walks Fine Line in Quest for White Votes: Campaign Curtails Stops in Black Communities.” Boston Globe, May 1, A1.

INDEX

African American vote, 6, 11, 99, 100, 102, 110–11, 115, 139nn5–6, 140nn15–16; early primaries, 118–21; Jackson and Obama compared, 132–37, 133; Jackson’s appeal to, 134–35, 135; Nevada, 112–18; primaries vs. caucuses, 122–25; race gap and voter preferences, 127–32; South Carolina, 112–18, 131–32 AIG, 67 Alabama primary, 118 Alito, Samuel, 48 American Issues Project (AIP), 61 American National Election Studies (ANES) surveys, 97 AON Risk Services, 67–68 attrition models, 103–4, 104 back-loading, 90 barometer equality, 39 Barone, Michael, 117 Bartels, Larry, 88 Bauer, Gary, 45n11 Biden, Joseph, 64, 103 Bipartisan Campaign Reform Act of 2002 (BCRA) (McCain-Feingold),

4, 28, 48, 49; criticisms of, 50–51; party vs. nonparty fundraising, 50; soft money limitations, 50–52. See also campaign finance bloggers, 2 Boehner, John, 64, 74n44 Bopp, James, Jr., 48, 50 Bradley, Bill, 17, 20, 102 Brooks, Robert C., 62 Brown, Jerry, 21, 97, 100 bundlers/bundling, 27, 30–31, 31, 35–37, 51; corruption and, 57–58; race and gender, role of, 126–27; reporting requirements, 45n9, 58–59, 72n6 Burroughs v. United States, 63 Busch, Andrew E., 20, 22, 121 Bush, George H. W., 97, 100 Bush, George W., 18, 28, 58, 83, 147; declines public financing, 35, 47, 51; microdonors and, 33, 34 California Fair Political Practices Commission, 71 California primary, 90 campaign committees, corruption and, 62–70

159

160

Index

campaign finance, 23; candidate attributes and quality, 41–42, 43–44; complexity of, 66–67; current status of primary fundraising, 37–38; early primaries and, 117, 117–18; electioneering communications, regulation of, 48–50; evolution of, 17–18; expenditures, 68–71; explaining changes, 34–38; frontloading and, 91–92; hard money, 51, 53, 55; internal management, 58; limits on individual contributions, 28, 30, 35–36, 39, 43, 49, 51, 78, 92, 112, 125; maxed-out donors, 37, 40, 42, 45n18; money supply, 35, 37, 43; 1974 reforms, 28, 47, 78; normative evaluation of trends, 38–42; political equality, 39–41, 43; rise of bundling, microdonations, and microdonors, 35–37; rules and dynamic change, 28–34; small donors, 2, 17–18, 20–21; soft money donations, 36, 49, 50–52; total contributions, 31–32, 32, 44n8, 60; total money raised by all candidates, 29, 30; transaction costs, 37, 54; 2004 nomination process, 28–29; voluntary partial public, 2, 4, 17–18. See also Bipartisan Campaign Reform Act of 2002; bundlers/bundling; corruption; Internet fundraising; McGovern-Fraser reforms; microdonations; microdonors; public financing system; reporting requirements

Campaign Finance Institute, 37, 60 candidates: advocacy/issue, 41, 81, 82, 104, 105, 122; attributes and quality, 41–42, 43–44, 98–99; third-party, 97 Capps, Lois, 64, 74n44 Carter, Jimmy, 7, 87–88, 99 caucuses, 7, 9, 24–25n5, 91, 111, 114, 122–25. See also Iowa caucuses Center for Responsive Politics, 70 Chavez, Linda, 69 cheap speech, 37, 42, 43 Citizens for Responsibility and Ethics in Washington, 70 Clinton, Bill, 98 Clinton, Hillary, 2, 8; African American vote and, 99, 119–20; bundled funders, 127, 139n12; caucuses and, 9; declines public financing, 35, 47, 51–52, 112, 125–26; endorses Obama, 98; entry fee, 80; expenditures, 69; fundraising, 18, 45nn14, 18, 52; gender gap and, 100; gender identity comments about, 132; microdonations to, 51, 60; Missouri primaries and, 11, 11; pledged delegate allocation and, 12; prediction models and, 105; race and gender in campaign finance, 125–27; South Carolina vote and, 116; superdelegates and, 14; withdrawal from race, 82, 83, 87 Clinton-Obama ticket, 16 collective action problem, 16–17 Colorado primary, 120

Index Commission on Presidential Nomination Timing and Scheduling, 112, 143, 144–45 congressional districts, 10 Connally, John, 18, 44n6 constituency, politician’s, 127 context-specific contests, 20 corruption, 57–58; campaign committees and, 62–70; duty driven, 62; expenditures and, 68–71; fraudulent audits, 63–64; uses of political funds, 71–72. See also campaign finance D. H. Lloyd & Associates, 68 Dean, Howard, 16, 17, 18, 45n12, 139n11; declines public financing, 35, 47, 51; on early primaries, 113; party control and, 19 Delaware plan, 93n2, 147 delegate allocation, 2–3, 77, 78; bonus delegates, 144; docking of for violation of windows, 85, 91–92, 110–11, 119–20, 145; front-loading trend and, 3, 85–86; pledged, 12, 12–14, 13; primary vs. caucus mode, 122–25, 123; superdelegates, 6, 13–17, 85–86, 107n4, 143–44, 147; winner-takeall vs. proportionality, 7, 10–13. See also front-loading Democratic National Committee, 14 Democratic National Convention, 1968, 15, 24n1, 143 Democratic National Convention, 2004: Commission on Presidential Nomination Timing and Scheduling, 112, 143, 144–45

161

Democratic National Convention, 2008, 22, 89, 98; Democratic Change Commission, 22, 89, 146–47 Democratic Party: campaign finance issues, 37–38; delegate allocation, 85–86, 111; gender gap, 101; proportionality and, 10–13; Rules and Bylaws Committee (RBC), 112–13, 146; superdelegates and, 13–17; voter participation, 102 Democratic Senatorial Campaign Committee (DSCC), 63–64 descriptive representation, 113–14 district conventions, 5 Dodd, Christopher, 81, 103 Doggett, Lloyd, 64, 74n44 Dole, Elizabeth, 64 Dole, Robert (“Bob”), 20, 83, 100 donors, demographics, 39–41. See also bundlers/bundling; microdonations; microdonors Dukakis, Michael, 134, 135, 140n14 Edwards, John, 11, 12, 12, 29, 70, 81, 82, 104; front-loading and, 119; race and gender, role of, 128; South Carolina vote and, 116 electioneering communications, regulation of, 48–50, 52, 60–62 end-game predictors, 103–4, 105 Eppard, Ann, 65 Era of Good Feelings, 24n3 factional candidates, 8 family employees, 70–71 FEC v.Wisconsin Right to Life, 50

162

Index

Federal Election Campaign Act (FECA), 28, 47 Federal Election Commission (FEC), 28, 48, 58, 66, 74n37 female candidates, 21–22 Fenno, Richard, 127 Ferraro, Geraldine, 131 First Amendment, 38, 50 Florida primary, 2, 22, 91, 111; gender and race, role of, 119–20; Latino vote, 120–21 Forbes, Malcolm (“Steve”), Jr., 18, 44n6 Ford, Gerald, 7–8, 87–88 foreign nationals, 60 Fraser, Donald, 24n1 front-loading, 20, 92–93, 141, 145–46; assumptions about, 79–80, 79–87; campaign finance revision, 91–92; complacency and inaction, 89–90; compression of calendar, 121–22; delegates as formality, 80, 85–86; dilution effect, 118–21; early decision and nonmeaningful participation, 79–80, 82–84, 84; early thinning of candidate field, 79, 81–82, 82; entry fee, 79, 80–81, 81; gender and race issues, 112–21; general election campaign length, 80, 86–87, 87; generalizations, 87–89; holding the line on windows, 90–91; incentives to move states back, 90; McGovern-Fraser reforms and, 110 front-runners, 7–9 Gallup polls, 102, 103 gender, role of, 109–10; front-loading, 112–22; gender gap, 98–102, 128,

129; primary vs. caucus, 122–25, 123; representation, 113–16; small donors and bundling, 126–27; voter preferences and, 127–32. See also race, role of generalizations, 92–93 Georgia primary, 118–19 Giuliani, Rudolph, 18, 74n42, 81, 85, 103 Glenn, John, 100 Gore, Al, 20, 47, 102, 140n14; African American vote and, 99; microdonations to, 32; public funds, use of, 28, 34 Graham, Lindsey, 64 Hanleigh Insurance, 67 Hart, Gary, 15, 19, 82 Hasen, Richard, 17, 23 Hayward, Allison, 17 Herman, Alexis, 142 Holbrook, Thomas, 102 House campaign accounts, 64 Huckabee, Mike, 11, 11, 13, 13, 18, 83, 104, 109–10; entry fee, 80; exit polls, 122; gender gap and, 100 Humphrey, Hubert, 6, 15, 97 Hunt, Jim, 143 Hunt Commission (1981–1982), 143–44 Hunter, Duncan, 81, 104 hybrid ads, 52 identity politics, 109–10 Indiana primary, 124, 139n6, 141 Inouye, Daniel, 64 insurance, liability, 67–68 Internet fundraising, 2, 4, 17–18, 20–21, 27, 51, 53–54; cheap

Index speech, 37, 42, 43; e-mail, 36; microdonations and microdonors, 36–37; reporting requirements, 59–60; social networking sites, 36; text messaging, 36 intraparty competition, 4–5 invisible primary, 2, 3, 4, 42, 43, 79 Iowa caucuses, 2, 3, 6–7, 9, 16–17, 19, 25n7, 78, 84 Iranian hostage crisis, 87–88 issue advertising, 60–62 Jackson, Andrew, 5 Jackson, Jesse, 15, 97, 99, 131, 140n14; Obama, comparison with, 132–37, 133; white support for, 129, 132–34 joint fundraising committees, 51–52, 66 Kennedy, Edward, 25n6, 87–88, 97 Kennedy, John F., 6, 13, 56, 97 Kerry, John, 18, 47, 83, 84; African American vote and, 99; declines public financing, 35, 51; microdonations to, 32; microdonors and, 33, 33, 34; party control and, 19 Keyes, Alan, 45n11, 81 King Caucus, 5 Kucinich, Dennis, 45n11, 81, 104 LaRouche, Lyndon, 69 Latino vote, 11, 100, 109, 111, 113–16, 118, 120–21, 124–25, 128, 131, 137, 139n9 leadership PACs, 66 Lodge, Henry Cabot, 63 Lott, Trent, 65

163

major purpose standard, 61 Mann, Thomas, 17, 23 Marshall, Thomas R., 123–24 Massachusetts Citizens for Life, 60–61 Mayer, William G., 83, 121 McAuliffe, Terry, 112 McCain, John, 2, 5, 9, 11, 11, 17–18; bundled funds, 37, 45n16; declines public financing, 35, 47; early decision and, 83–84, 91n1; entry fee, 80–81; expenditures, 69, 74n41; front-loading and, 80–81, 83–87, 120–21; gender gap and, 100; microdonations to, 60; microdonors and, 33, 34, 51; private contributions to, 29; public financing and, 29, 35, 44n5, 47, 52–53; Super Tuesday and, 13, 13; winner-take-all and, 25n8, 121 McCain-Feingold Act. See Bipartisan Campaign Reform Act of 2002 McGovern, George, 7, 24–25n5, 100 McGovern campaign, 2–3 McGovern-Fraser Commission, 143 McGovern-Fraser reforms, 1, 7–8, 14–15, 24n1, 110–12 media, 1, 4, 7, 116–17 Michigan primary, 2, 22, 85, 91, 110–11, 113, 144; gender and race, role of, 119–20, 124 microdonations, 27, 29, 31–34, 32, 33, 35–37; relative value of, 125–26 microdonors, 31, 32–33, 33, 35–37, 45nn13, 18; race and gender, role of, 125–26 Mikulski Commission, 143 minority empowerment thesis, 114–15

164

Index

Missouri, 10–11, 11 Mondale, Walter, 15, 20, 25n12, 83, 86 Monroe, James, 24n3 Muskie, Edmund, 7, 97 national conventions, 5, 142 national primary concept, 141–43 National Republican Congressional Committee (NRCC), 63, 65 Nevada primary, 110–11, 112–18, 139n9, 144; representation, 113–16 New Hampshire primary, 2, 3, 6, 78, 78; McCain and, 83–84 Nixon, Richard M., 7 nomination process: change and continuity, 1–5; democratization of, 22–23, 109; lessons of 2008, 7–18; political parties in, 5–7; state-by-state sequence, 142 nonprofits, qualified, 60–62 Norrander, Barbara, 3 North Carolina primary, 124, 139n6, 141 Northup, Anne, 65 Obama, Barack, 13–14, 18; African American vote and, 119–20; bundled funds, 36–37, 45n16, 127, 139n12; declines public financing, 18, 35, 47, 51–53, 92, 112, 125–26; downplays race issues, 131, 132–37; entry fee, 80; expenditures, 69, 74n40; frontloading and, 86–87, 119; individual donors, 45n13; Jackson, comparison with, 132–37, 133; as keynote speaker, 2004, 8; microdonations to, 60;

microdonors and, 32–33, 33, 34, 45n13, 51, 59–60; Missouri primaries and, 11, 11; pledged delegate allocation and, 12, 12; prediction models and, 105; private contributions to, 29; public financing and, 44n5, 47–48; race and gender in campaign finance, 125–27; South Carolina vote and, 116–17; white support for, 129, 132–34 Ohio Plan, 93n2, 147 outside groups, 20–21, 48–49, 52 Palin, Sarah, 5, 75n52 Party Leader/Elected Official delegates (PL/EOs). See superdelegates Paul, Ron, 9, 11, 18, 45n11, 69, 81, 100, 104 Pennsylvania primary, 132 “Pioneers” (Bush bundlers), 30 Pitkin, Hannah Fenichel, 114 pledged delegate allocation, 12, 12–14, 13 political equality, 39–41, 43 political market failure, 146 political parties: leadership, 8, 15–16; in nominations process, 5–7; party control, 19–21; permeability of, 8–9 Political Pulse:The Newsletter of California Politics and Government, 90 Polsby, Nelson W., 8 Powell, Colin, 21–22 prediction: attrition models, 103–4, 104; end-game predictors, 103–4, 105; funding models, 102–3

Index Presidential Primaries and the Dynamics of Public Choice (Bartels), 88 Price, David, 14, 15 primaries, 7; candidate’s appeal and, 6; caucuses vs., 111, 122–25, 123; evolution of, 6, 95; gender and race issues, 118–21; gender gap, 2000, 101; gender gap, 2008, 101, 101–2; invisible, 2, 3, 4, 42, 43; McGovern-Fraser reforms, 110–11; national primary concept, 141–43; as non-divisive, 95–98; predictions, 102–6; pre-window period, 85, 90–91, 110–13, 116, 118–21, 123, 127; regional, 23; representation, 113–16; shaping course of subsequent campaign, 116–18; voters and, 96–97. See also front-loading; Super Tuesday; specific primaries primary season: campaign finance and, 28–29, 32, 34–35, 37, 39, 42, 45n16; front-loading and, 83; race and gender issues, 106, 117, 134, 139n6 proportionality, 3, 7, 10–13, 12, 13 public financing system: demise of, 34–35; 52 million spending cap, 28, 29, 103; nomination component, 48; public matching funds, 23, 28, 34, 39, 43, 47, 51–55, 125–26; rules, 28; sign of weakness to opt into, 47. See also campaign finance public opinion polls, 8 qualified nonprofit, 60–62 race, role of, 109–10; front-loading, 112–22; group identity, decline

165

in, 135; Jackson and Obama compared, 132–37, 133; minority empowerment thesis, 114–15; primary vs. caucus, 122–25, 123; race gap, 129; representation, 113–16; voter preferences and, 127–32. See also gender, role of Reagan, Ronald, 7–8, 87–88 reform commissions, 22, 89–90 regional primaries, 23 Reiter, Howard, 24n4 reporting requirements: bundlers/bundling, 45n9, 58–59, 72n6; electioneering communications, 60–62; expenditures and, 68–71; filing and review, 67; individual donors and, 68, 70; Internet fundraising, 59–60. See also campaign finance Republican National Committee, 25n10, 53, 147; Temporary Delegate Selection Committee, 22, 89, 147 Republican National Convention, 89, 147 Republican Party: campaign finance issues, 37–38; delegates and, 85, 111; gender gap in voting, 99–100, 101; McGovern-Fraser reforms and, 7–8; proportionality and, 10 Richardson, Bill, 81, 109, 115, 127 Roberts, John, 38, 53 Robertson, Pat, 21 robot rule, 25n6 Romney, Mitt, 9, 11, 11, 13, 13, 18, 41, 100, 103, 109–10; entry fee, 80; front-loading and, 81, 84, 85, 121 Roosevelt, Theodore, 95

166

Index

Rove, Karl, 147 Rumsfeld, Donald, 57 Russert, Tim, 124

Texas Plan, 93n2 Thompson, Fred, 81, 93n1, 104 Tsongas, Paul, 64, 100, 104

self-financed candidates, 28 Senate campaign accounts, 63–64 Sharpton, Al, 99 Shuster, Bud, 65 Smith, Al, 63 Smith, Bradley, 48 South Carolina primary, 3, 85, 110–18, 137, 144; African American vote, 131–32; representation, 113–16 Spakovsky, Hans von, 48 star persona, 41–42, 43–44 state conventions, 3, 5 Steen, Jennifer A., 11, 21 Stevenson, Adlai, 6 substantive representation, 114 superdelegates, 6, 13–17, 85–86, 91, 107n4, 143–44, 147 superdonors, 38 Super Tuesday, 10, 13, 13, 83, 86, 140n14; race and gender, role of, 118, 120, 138 Supreme Court, 38, 52, 53, 60–61

unintended consequences, law of, 141

Taft, William, 95 Tancredo, Tom, 45n11, 81

Vietnam War, 15 voter fatigue, 86 voter participation, 2, 10; black turnout, 114–15, 138n4, 139n5; Nevada and South Carolina primaries, 114–16 voters, rank and file, 1, 4–5, 14; identity groups, 109–10; primaries and, 96–97; race and gender, role of, 127–32. See also African American vote; gender, role of; Latino vote Wachovia Insurance Services, 67 Wallace, George, 21, 97 war, nominations process and, 20 Ward, Christopher, 63, 65 West Virginia primary, 6, 136 Willkie, Wendell, 6 winner-take-all, 7, 10–13, 24, 25n8, 90–91, 121 Winograd Commission, 143 women voters. See gender, role of Wright, Jeremiah, 131–32

ABOUT THE EDITORS AND CONTRIBUTORS

Jack Citrin is the Heller Professor of Political Science and the director of the Institute of Governmental Studies at the University of California, Berkeley. David Karol is an assistant professor of political science at the University of California, Berkeley.

h Andrew E. Busch is a professor of government at Claremont McKenna College. Richard L. Hasen is the William H. Hannon Distinguished Professor of Law at Loyola Law School. Allison R. Hayward is an assistant professor of law at George Mason School of Law. Thomas E. Mann is the W. Averell Harriman Chair and Senior Fellow in Governance Studies at the Brookings Institution. Barbara Norrander is a professor of political science at the University of Arizona. David E. Price is the U.S. Representative for North Carolina’s 4th District and a professor of political science and public policy at Duke University. Jennifer A. Steen is a lecturer in political science at Yale University. 167